{"as_of":"2026-08-19T00:52:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:9083087a66135fac264e420c8315bf7ceae971a7f62406aed34e66aa1de6e9cf","coverage":[{"denominator":147,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":100,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-15T19:24:57.553636Z","state":"measured"},{"denominator":100,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":100,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-18T06:34:40.430872+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2506.16790/citation-record","integrity":"/paper/2506.16790/integrity","json":"/paper/2506.16790/citation-record.json","paper":"/paper/2506.16790"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.125450Z","title":"Mixhop : Higher-order graph convolutional architectures via sparsified neighborhood mixing","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.125450Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:8c013efce7845fabe52f67d22699891b288202b65a58000429393b10fcc1bec5","observation_id":"c91fc1ca-4ea5-40ac-9448-e45d6ba2d816","resolution":{"observed_at":"2026-08-15T19:24:57.125450Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.130809Z","title":"On the bottleneck of graph neural networks and its practical implications","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.130809Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:9537d18f12faea6ea248accd7c1333ab892edbfb6f42b504b5ed9ce0ed8d61c9","observation_id":"a52379f8-dc21-4993-830f-0988e8c5d4f6","resolution":{"observed_at":"2026-08-15T19:24:57.130809Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.135531Z","title":"A machine learning-based approximation of strong branching","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.135531Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:cdca47fe608cb772410454d9a7fd1fe54bff33af9dc7488a3f6dcf73dce63453","observation_id":"fd969f21-f063-4ef4-a6bc-f2c6d66d49a5","resolution":{"observed_at":"2026-08-15T19:24:57.135531Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.140228Z","title":"Label propagation across graphs: Node classification using graph neural tangent kernels","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.140228Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:6dc0a0f615840c059f5614289e91b76238e55a29658b2e1e74ebc39af8490685","observation_id":"2e828d5d-d28b-4df0-a410-c057ec0dff3a","resolution":{"observed_at":"2026-08-15T19:24:57.140228Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.144838Z","title":"Machine learning for combinatorial optimization: a methodological tour d’horizon","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.144838Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:24743dc20dd23f5b7134502693c99c9f6f30ac5ddd44f4e7fbbfc07903c8abc7","observation_id":"f68589fe-2721-4a7f-b982-8e87edd73a6c","resolution":{"observed_at":"2026-08-15T19:24:57.144838Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.148857Z","title":"Decision diagrams for optimization, volume 1","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.148857Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:950c6e469f6ad3dd1fde6cb6bdb4ed066479218691ed9f692b9c3a4cd5d3fb2a","observation_id":"a04667d6-c75c-4feb-a006-d98c8a2f6655","resolution":{"observed_at":"2026-08-15T19:24:57.148857Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.152982Z","title":"The maximum clique problem","venue":null,"work_id":null,"year":1999},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.152982Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:3313b9eee5d425738a7642c9cba906862c085688298b6c6dc304246d303a339f","observation_id":"1ea5bf44-2c9c-4e1c-9040-9c01e1010c1e","resolution":{"observed_at":"2026-08-15T19:24:57.152982Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.157230Z","title":"Can graph neural networks go deeper without over-smoothing? yes, with a randomized path exploration! In IEEE Transactions on Emerging Topics in Computational Intelligence","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.157230Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:41028095ff439f8cad7fb6677129ece1178d3b7df2d4c76c4524194a968c5a97","observation_id":"e053edc7-435d-4efa-98a7-77ba6f07dadb","resolution":{"observed_at":"2026-08-15T19:24:57.157230Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.161303Z","title":"Reconnaissance de la parole par reseaux connexionnistes","venue":null,"work_id":null,"year":1988},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.161303Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:1174acb1e7055469bf51cbd1c8cb04da8ec4f62021fc8fa3fd6329ad40c4d63c","observation_id":"885cb1ff-2ffa-4961-99ab-793aa64adfc7","resolution":{"observed_at":"2026-08-15T19:24:57.161303Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.165310Z","title":"A survey on optimization metaheuristics","venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.165310Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:7b9fc42d1264e6c8d51f5b211a071929e985391ecaf78f1fc0b2b12d5b0ece54","observation_id":"722919a5-1b5d-4a2a-9543-99b1138aa357","resolution":{"observed_at":"2026-08-15T19:24:57.165310Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.169064Z","title":"A note on over-smoothing for graph neural networks","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.169064Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:196a18c1cd6725f3680adb255989694515d7ea7b007daa22191409ec8aa119a0","observation_id":"2dc6f21a-f8a1-4ad6-9540-98f22304748e","resolution":{"observed_at":"2026-08-15T19:24:57.169064Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.173370Z","title":"Measuring and relieving the over-smoothing problem for graph neural networks from the topological view","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.173370Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:db8df4d48d6670cd487f2511032b79b95c6c29e3963e034dbe0618aafe2d8b10","observation_id":"836d7a64-8228-41f0-ab1e-5dfc691939dd","resolution":{"observed_at":"2026-08-15T19:24:57.173370Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.178559Z","title":"Dirichlet energy enhancement of graph neural networks by framelet augmentation","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.178559Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:7aaceb7403696224c44a85546730dfbaee4878bbbc7921833c3132ae584885e0","observation_id":"33613da8-c84f-4000-81c1-b09d091f8048","resolution":{"observed_at":"2026-08-15T19:24:57.178559Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2205.15127","last_updated":"2022-05-30T14:19:45Z","snapshot_observed_at":"2026-08-16T16:56:39.005814Z","submitted_at":"2022-05-30T14:19:45Z","title":"Universal Deep GNNs: Rethinking Residual Connection in GNNs from a Path Decomposition Perspective for Preventing the Over-smoothing","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2205.15127","snapshot_observed_at":"2026-08-15T19:24:57.183603Z","title":"Universal deep gnns: Rethinking residual connection in gnns from a path decomposition perspective for preventing the over-smoothing","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.183603Z"},"links":{"cited_paper":"/paper/2205.15127","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:c6e62c3e1c480d7aab80669361575dcdae528a5e30240623f0345351db147f60","observation_id":"965abd38-b443-4293-9d68-374e3f5f63cf","resolution":{"observed_at":"2026-08-15T19:24:57.183603Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.188886Z","title":"Simple and deep graph convolutional networks","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.188886Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:d9a75d6693fc25d54491696ccd0565f8f125904175f0acfa43935058077beddd","observation_id":"79c4674c-593e-475d-96e1-95e868a8d056","resolution":{"observed_at":"2026-08-15T19:24:57.188886Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.193479Z","title":"Dynamical isometry and a mean field theory of rnns: Gating enables signal propagation in recurrent neural networks","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.193479Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:19175471f347d8dc48d19cb5f55e82f4c6a0cc153d68b78bce5b4e852276fa09","observation_id":"8e9afcfb-5412-48e8-9353-a5f627c57b31","resolution":{"observed_at":"2026-08-15T19:24:57.193479Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.198105Z","title":"Bag of tricks for training deeper graph neural networks: A comprehensive benchmark study","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.198105Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:aba3f9059293ff6829ae523336b2d1fd2da6c529e00981cc4039d63a08c9a1ae","observation_id":"3dfdf6cc-0ddc-4e71-b30b-c4f54f7ff572","resolution":{"observed_at":"2026-08-15T19:24:57.198105Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.203774Z","title":"On representing linear programs by graph neural networks","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.203774Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:e9aa45d330b31e1b8c714774b6aed5dd6d591cc100944524fd207c5c0549f4e7","observation_id":"5fbbb7bd-41eb-44be-917b-88e0c5acc06e","resolution":{"observed_at":"2026-08-15T19:24:57.203774Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.208555Z","title":"Expressive power of graph neural networks for (mixed-integer) quadratic programs","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.208555Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:d3476b07e75968aa977f18de5701fd60da99f509331ddc1c0b3f0336d0d15a92","observation_id":"41affb3d-1773-4747-ae36-d81d325aa57f","resolution":{"observed_at":"2026-08-15T19:24:57.208555Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.212766Z","title":"Adaptive universal generalized pagerank graph neural network","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.212766Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:f2ab4d79baca391ad38f16e7c91b5f8da2369302ab0d266e71f9c18022508bd7","observation_id":"5d434ef5-ebd9-49b9-89bd-d47077b03b58","resolution":{"observed_at":"2026-08-15T19:24:57.212766Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2408.04895","last_updated":"2024-11-02T06:10:08Z","snapshot_observed_at":"2026-08-17T16:06:43.692790Z","submitted_at":"2024-08-09T06:46:06Z","title":"Better Not to Propagate: Understanding Edge Uncertainty and Over-smoothing in Signed Graph Neural Networks","version":3},"cited_work":{"arxiv_id":"2408.04895","doi":null,"metadata_source":"pith","pith_arxiv_id":"2408.04895","snapshot_observed_at":"2026-08-15T19:24:58.367439Z","title":"Better Not to Propagate: Understanding Edge Uncertainty and Over-smoothing in Signed Graph Neural Networks","venue":"cs.LG","work_id":"298880ad-51d9-4086-b946-d51d906fac24","year":2024},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.216873Z"},"links":{"cited_paper":"/paper/2408.04895","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:9c0f3175b31b7b18b1a793bf22e88d80ebaa411d17711eb4440c997ba37b7b3a","observation_id":"1c0385a8-57e2-49f1-9d77-398696756b45","resolution":{"observed_at":"2026-08-15T19:24:58.371978Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.221394Z","title":"Approximation algorithms for bin-packing—an updated survey","venue":null,"work_id":null,"year":1984},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.221394Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:83c42e7e60b0b1087613015453201023620bb285104f6de0b1a488b2a66ce01d","observation_id":"7e08e607-16ac-4cb2-ac43-b0da85090cb6","resolution":{"observed_at":"2026-08-15T19:24:57.221394Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.225961Z","title":"On provable benefits of depth in training graph convolutional networks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.225961Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:455d90a056d5dff1afd26c26c95b66c03c1776a0713bf319f84d6f162a4eac0a","observation_id":"2a89a58c-630a-4322-ac15-49342cf95d16","resolution":{"observed_at":"2026-08-15T19:24:57.225961Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.229567Z","title":"Metainit: Initializing learning by learning to initialize","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.229567Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:fd161c2ae9c453ba53fa1436833e4b36f5c816892d302c2d23c190d61a640677","observation_id":"ac92d969-592b-4c31-80e1-5b4af06ca604","resolution":{"observed_at":"2026-08-15T19:24:57.229567Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2206.10991","last_updated":"2023-09-06T06:58:11Z","snapshot_observed_at":"2026-08-16T22:36:36.439581Z","submitted_at":"2022-06-22T11:45:36Z","title":"Understanding convolution on graphs via energies","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2206.10991","snapshot_observed_at":"2026-08-15T19:24:57.233344Z","title":"Graph neural networks as gradient flows: Understanding graph convolutions via energy","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.233344Z"},"links":{"cited_paper":"/paper/2206.10991","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:297d87155fde604ab9cc7e3a658bc7aabaaa8b17eca82f2e54da2f584db25300","observation_id":"60be6f7d-6686-4b3f-9be8-10d200ce0028","resolution":{"observed_at":"2026-08-15T19:24:57.233344Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.237971Z","title":"Adagnn: Graph neural networks with adaptive frequency response filter","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.237971Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:65c0ab62df38e3de1993650573eb1516427540a4b6d7e544aad72ceae1c14b2b","observation_id":"45e510d2-0f93-48ca-8eb6-7ad1df7b5c5c","resolution":{"observed_at":"2026-08-15T19:24:57.237971Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.241791Z","title":"Graph neural tangent kernel: Fusing graph neural networks with graph kernels","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.241791Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:c23c1220f991949fb40421ec15cd870007a52281bde30c7db69c06b5f115d19b","observation_id":"31e9114a-7cf5-420d-a082-f2f9cd8d781f","resolution":{"observed_at":"2026-08-15T19:24:57.241791Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.245697Z","title":"Learning from the dark: boosting graph convolutional neural networks with diverse negative samples","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.245697Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:bd08bf0bde89f9f61cfb48b490634fbc0720468c3705650dd326e429ece764be","observation_id":"ac014f7d-d628-4c96-824c-075533f6ac38","resolution":{"observed_at":"2026-08-15T19:24:57.245697Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.249396Z","title":"Graph convolutional neural networks with diverse negative samples via decomposed determinant point processes","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.249396Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:33d3f27594ff0ec63e4be1b05899bd90f8366534bcb5e26bfe9ad4c791069f2d","observation_id":"8c1ad13c-9d26-4553-b872-484b09b846fb","resolution":{"observed_at":"2026-08-15T19:24:57.249396Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.253225Z","title":"Layer-diverse negative sampling for graph neural networks","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.253225Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:f84dd544ac22b83aece1381de5e7a68eff8b87cf4f220c55baf476273892c8e5","observation_id":"621c0bb0-0d8d-45ec-bae7-84e219b48b18","resolution":{"observed_at":"2026-08-15T19:24:57.253225Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.257176Z","title":"The power of depth for feedforward neural networks","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.257176Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:5ba2f793d9f66ceda4ad3d9bdf174735ddf868ee492c3785d525cbf7020541ca","observation_id":"3656e498-d58b-495e-a0e8-dc9caaab6ca5","resolution":{"observed_at":"2026-08-15T19:24:57.257176Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.260990Z","title":"DropMessage : Unifying random dropping for graph neural networks","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.260990Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:de5a934275a35fa9e282c825c6783354f8d29d49def2a2f3881641138494dd7d","observation_id":"e6ba5654-c20a-4c5a-8873-961247db9537","resolution":{"observed_at":"2026-08-15T19:24:57.260990Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.265198Z","title":"Grato: Graph neural network framework tackling over-smoothing with neural architecture search","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.265198Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:adf9a70da76514a0f7a9c9df2604b460aba99f4030585ad38aff1aef9cf1b613","observation_id":"9adbaf49-5938-46a4-bcc1-b51b825b70cd","resolution":{"observed_at":"2026-08-15T19:24:57.265198Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.270059Z","title":null,"venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.270059Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:9feb450c082c061947b3d6c3ed2362503ce7a543fd5199336a374809241eaf70","observation_id":"b86f95be-218a-49dc-8f7e-acd537e9302e","resolution":{"observed_at":"2026-08-15T19:24:57.270059Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.01267","last_updated":"2024-06-09T01:42:46Z","snapshot_observed_at":"2026-08-16T14:55:49.748264Z","submitted_at":"2023-10-02T15:08:52Z","title":"Cooperative Graph Neural Networks","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.01267","snapshot_observed_at":"2026-08-15T19:24:57.274279Z","title":"Cooperative graph neural networks","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.274279Z"},"links":{"cited_paper":"/paper/2310.01267","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:c21c7fb917fa3af67f11850bd0f4b8bef763b4733e237f304d00e56a5f6519e0","observation_id":"e108765b-40d5-4578-a6e0-39189c8262e9","resolution":{"observed_at":"2026-08-15T19:24:57.274279Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.278366Z","title":"Exact combinatorial optimization with graph convolutional neural networks","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.278366Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:31c1c783d62682daee7dbfffb9ca6b0ab7087afb3ae1d88c908f2e9ef0931d67","observation_id":"09068c51-fd1a-46a4-8809-1e63687ecab6","resolution":{"observed_at":"2026-08-15T19:24:57.278366Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.283704Z","title":"Predict then propagate: Graph neural networks meet personalized pagerank","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.283704Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:7e84442ea98419768e8c299ef77976fee8f0a6fcf66ebe3897d4bffb6248d0fe","observation_id":"3651b512-aec5-43ea-af88-41d225370c0f","resolution":{"observed_at":"2026-08-15T19:24:57.283704Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.287779Z","title":"The travelling salesman problem and related problems","venue":null,"work_id":null,"year":1978},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":38,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.287779Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:a236bb60f72c2dd5993ccda71fd48015aae92f557d309eed7750782cb7bed8e1","observation_id":"ac9e7fe6-52a9-4fac-a681-e6763f020b7c","resolution":{"observed_at":"2026-08-15T19:24:57.287779Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.291486Z","title":"Graph convolutional networks from the perspective of sheaves and the neural tangent kernel","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.291486Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:173cf3f2af9f6a8a094e53cf7787cca4da7fe42945c3a68b7c0c1a62c5ba83c3","observation_id":"7eb00f4a-6ef1-484d-b288-3dddf1fdc8cd","resolution":{"observed_at":"2026-08-15T19:24:57.291486Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1901.08987","last_updated":"2019-05-24T01:10:22Z","snapshot_observed_at":"2026-08-14T17:25:50.300815Z","submitted_at":"2019-01-25T17:05:54Z","title":"Dynamical Isometry and a Mean Field Theory of LSTMs and GRUs","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1901.08987","snapshot_observed_at":"2026-08-15T19:24:57.295398Z","title":"Dynamical isometry and a mean field theory of lstms and grus","venue":null,"work_id":null,"year":1901},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.295398Z"},"links":{"cited_paper":"/paper/1901.08987","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:7d97fb619082c45ba380aa4a205b6b7eec756663b00170c36b294db5688d333b","observation_id":"6db5fc30-5048-4951-8af5-dbf317370da1","resolution":{"observed_at":"2026-08-15T19:24:57.295398Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.299939Z","title":"Understanding the difficulty of training deep feedforward neural networks","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.299939Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:6e42a38fbff477e6001e76fdf87d262683af964b966f3fada2aa86d421ec45ea","observation_id":"5ca7c281-f2a1-46b2-a255-f18ecc47f2a6","resolution":{"observed_at":"2026-08-15T19:24:57.299939Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.304468Z","title":"Orthogonal graph neural networks","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":42,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.304468Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:76e3d195b361f01e983178d690bb41987b7fd40891f4289e4824f213cd71fd3e","observation_id":"16ee122a-9cbd-4e15-a6bc-0351ecea1e7f","resolution":{"observed_at":"2026-08-15T19:24:57.304468Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.308345Z","title":"Contranorm: A contrastive learning perspective on oversmoothing and beyond","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":43,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.308345Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:6dea1ddfbbffd4c160fd19e5c10723ff1413097d4d25b84c29cfeccbfc9a8c54","observation_id":"ef33b669-d633-4d61-a10f-4f30e59efe82","resolution":{"observed_at":"2026-08-15T19:24:57.308345Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.312707Z","title":"Structure-aware dropedge toward deep graph convolutional networks","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":44,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.312707Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:3589c90c5ecf9a2195a89b94aa0030ec53aae9e8a1a200418b248860ccdf9075","observation_id":"97ebbe23-c890-47a8-8329-806df2307d54","resolution":{"observed_at":"2026-08-15T19:24:57.312707Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.316760Z","title":"A gnn-guided predict-and-search framework for mixed-integer linear programming","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":45,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.316760Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:0184c0fd9ba6e676cd00c26acd1a2f5bb29343b278a16f2717762c1f2d66a8d3","observation_id":"0c34132f-96b4-477f-b447-9a0317465f16","resolution":{"observed_at":"2026-08-15T19:24:57.316760Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.320668Z","title":"Mlpinit: Embarrassingly simple gnn training acceleration with mlp initialization","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.320668Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:2da3180a627a57bb0d21d2b1d8508c9d3d15205b8495b4f031f5a01d932700cc","observation_id":"ccbb12d0-971d-4dde-a8c3-234a8923b746","resolution":{"observed_at":"2026-08-15T19:24:57.320668Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.325302Z","title":"Which neural net architectures give rise to exploding and vanishing gradients? Advances in Neural Information Processing Systems, 31, 2018","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":47,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.325302Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:36c7f77d4e323929a91bea1fb442c112105c3c77641e5375e0b5b8236e9d4d90","observation_id":"64449fb6-085d-422e-ae78-fe0b6fdf0e2a","resolution":{"observed_at":"2026-08-15T19:24:57.325302Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.329349Z","title":"Inequalities","venue":null,"work_id":null,"year":1952},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":48,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.329349Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:c1a24d6d02275e61fd23be96b9b95ea8a247e408041bcee562e939b522ec42a3","observation_id":"715c994a-279b-46f7-a3d0-be6cb559b44d","resolution":{"observed_at":"2026-08-15T19:24:57.329349Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.333143Z","title":"On the impact of the activation function on deep neural networks training","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":49,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.333143Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:81c719c300a68126f24f04457cf3723120dcea5c23e172ea00b7fa71e80977bf","observation_id":"311bf995-afa0-4120-b21f-6564f9b3c9eb","resolution":{"observed_at":"2026-08-15T19:24:57.333143Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.337985Z","title":"The curse of depth in kernel regime","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":50,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.337985Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:05be1cc1e2ca4107d2e13d4435ef496193a32e8b60ad762810bbc12ef3244338","observation_id":"c6c48ed9-fde9-4ac0-acd4-17611cabe764","resolution":{"observed_at":"2026-08-15T19:24:57.337985Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.342413Z","title":"Delving deep into rectifiers: Surpassing human-level performance on imagenet classification","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":51,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.342413Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:cf00d0b2eb7a1d9d3db80a0222e5b6fae2cd58f2e1f7251a7053bd8148616fcc","observation_id":"7b8e1b15-bc81-4c63-a620-c8d7907dac51","resolution":{"observed_at":"2026-08-15T19:24:57.342413Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.346176Z","title":"Deep residual learning for image recognition","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":52,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.346176Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:248cd3b6d70959a6f66ee748f4504d0ded1c8bf83dd00254f700e3c929bc16f0","observation_id":"52bb6147-a673-4cf7-b1d1-0a4db78ae4a4","resolution":{"observed_at":"2026-08-15T19:24:57.346176Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.349790Z","title":"Open graph benchmark: Datasets for machine learning on graphs","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":53,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.349790Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:14e9c6086c77bfef7c59537583a61d55f458d06baab0a5656045d7c6416cf1e8","observation_id":"ccb84385-0851-4a09-97b2-452c0dcc1451","resolution":{"observed_at":"2026-08-15T19:24:57.349790Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.353217Z","title":"Towards deepening graph neural networks: A gntk-based optimization perspective","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":54,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.353217Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:3ef124a3e4757e5d70536bbe48193bea15df50c32237d49b062612fbb3c6c01a","observation_id":"971efa90-5585-45fa-ba8f-036325a644c2","resolution":{"observed_at":"2026-08-15T19:24:57.353217Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2008.09864","last_updated":"2022-07-09T01:41:01Z","snapshot_observed_at":"2026-08-07T03:30:20.016396Z","submitted_at":"2020-08-22T16:14:01Z","title":"Tackling Over-Smoothing for General Graph Convolutional Networks","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2008.09864","snapshot_observed_at":"2026-08-15T19:24:57.356848Z","title":"Tackling over-smoothing for general graph convolutional networks","venue":null,"work_id":null,"year":2008},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":55,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.356848Z"},"links":{"cited_paper":"/paper/2008.09864","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:aaa15605d1ff7092c19480a1a716fa0d80006c10ad96edb82eb9d08e8b9fac52","observation_id":"cdde0d09-08a9-4540-a33c-9b1ec448fe64","resolution":{"observed_at":"2026-08-15T19:24:57.356848Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.360839Z","title":"Batch normalization: Accelerating deep network training by reducing internal covariate shift","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":56,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.360839Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:e327bc553088c650153b2e74a45928055d20b8cf24a0d9165b1df25738e27fc1","observation_id":"5a3b771d-48b2-48ef-a254-280b2b2710c5","resolution":{"observed_at":"2026-08-15T19:24:57.360839Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.364607Z","title":"Old can be gold: Better gradient flow can make vanilla-gcns great again","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":57,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.364607Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:2c25f1da83c9b71f754f394bdf8ab6386f9de6101f9b5e15b948a0832ad257b9","observation_id":"eacffcf3-3007-4e26-84ea-a62f40ed271f","resolution":{"observed_at":"2026-08-15T19:24:57.364607Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.368784Z","title":"Fast graph neural tangent kernel via kronecker sketching","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":58,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.368784Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:e0aa0f1ff6050d3f7042511c1e8181bdb8e5a9cb17ad449cbac2ed79bb781e02","observation_id":"991eda1c-260d-41c0-8d87-d9d96af0ebd1","resolution":{"observed_at":"2026-08-15T19:24:57.368784Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.372827Z","title":"Towards feature overcorrelation in deeper graph neural networks","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":59,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.372827Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:df110f1e2b8e90c5a7b598ea2b4cc9cb3cd9b0f08cf6b284139076b1815732b4","observation_id":"f5ee31b8-3afb-45ae-b717-7671c4337684","resolution":{"observed_at":"2026-08-15T19:24:57.372827Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.377012Z","title":"Reducing oversmoothing in graph neural networks by changing the activation function","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":60,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.377012Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:8bebe7863bd2a69f11a88eff0a53b39fbc7621f9c883c139a7f085b8fd97c32f","observation_id":"7af2a450-8f55-4c1d-a80b-f016c51c3ad6","resolution":{"observed_at":"2026-08-15T19:24:57.377012Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.23830","last_updated":"2024-10-31T11:21:20Z","snapshot_observed_at":"2026-08-17T15:05:39.960827Z","submitted_at":"2024-10-31T11:21:20Z","title":"Reducing Oversmoothing through Informed Weight Initialization in Graph Neural Networks","version":1},"cited_work":{"arxiv_id":"2410.23830","doi":null,"metadata_source":"pith","pith_arxiv_id":"2410.23830","snapshot_observed_at":"2026-08-15T19:24:58.295715Z","title":"Reducing Oversmoothing through Informed Weight Initialization in Graph Neural Networks","venue":"cs.LG","work_id":"e0a1af21-4d57-4360-a774-3303f59ad7bf","year":2024},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":61,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.380565Z"},"links":{"cited_paper":"/paper/2410.23830","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:03bca93352c8fc4c5bdb7a451add350c35bd65a334ea8130a5cd9b57736601f9","observation_id":"8fd4e461-85c4-4018-99a1-9e3513db3c19","resolution":{"observed_at":"2026-08-15T19:24:58.301670Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.385708Z","title":"Not too little, not too much: a theoretical analysis of graph (over)smoothing","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":62,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.385708Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:b82cf81724041ec266d57ace1e8aa1d6abd8425ec96b7aaff33e43d2e8b14cbe","observation_id":"055a1984-9a77-40b8-bbe3-878948aebd87","resolution":{"observed_at":"2026-08-15T19:24:57.385708Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.389779Z","title":"Learning to branch in mixed integer programming","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":63,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.389779Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:84313cf56a9c38f31aab67a132a47902de57740c5d3b5ce9fce5806957c233bf","observation_id":"4b8f2f51-81a8-4324-afa9-4f2892898b60","resolution":{"observed_at":"2026-08-15T19:24:57.389779Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.393959Z","title":"Learning to run heuristics in tree search","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":64,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.393959Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:299ed9ebd37e0d056f4dcaf5ee48146afa1352f65104c49e8a21bd52515bfe4a","observation_id":"10e52646-e61d-4d99-8a55-0379419e2458","resolution":{"observed_at":"2026-08-15T19:24:57.393959Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.398296Z","title":"Semi-supervised classification with graph convolutional networks","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":65,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.398296Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:88ba4774d650fb797db48dd362a8593825bb4dd7d84bcecffccc9da3990b692d","observation_id":"449a0644-2b50-4eed-aa7b-94cc8762dac8","resolution":{"observed_at":"2026-08-15T19:24:57.398296Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.402705Z","title":"Goat: A global transformer on large-scale graphs","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":66,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.402705Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:4ad67d9c9de77395672e7992410cf8c281e323b731f7f00589b4a01866b0bef1","observation_id":"9ea9943b-5fd6-4908-91c6-4c765923f509","resolution":{"observed_at":"2026-08-15T19:24:57.402705Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2301.10808","last_updated":"2023-05-31T19:27:16Z","snapshot_observed_at":"2026-08-17T14:03:51.786463Z","submitted_at":"2023-01-25T19:52:58Z","title":"Graph Neural Tangent Kernel: Convergence on Large Graphs","version":2},"cited_work":{"arxiv_id":"2301.10808","doi":null,"metadata_source":"pith","pith_arxiv_id":"2301.10808","snapshot_observed_at":"2026-08-15T19:24:58.267552Z","title":"Graph Neural Tangent Kernel: Convergence on Large Graphs","venue":"cs.LG","work_id":"09e1a1bc-299d-4c01-9074-591c61318ee1","year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":67,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.406976Z"},"links":{"cited_paper":"/paper/2301.10808","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:e08e11848fede551d3ae45654870db737942ce26ce2e2cfca8c2f8ef6096b730","observation_id":"f48d6f55-59c6-4c3f-8c85-382115fb30ee","resolution":{"observed_at":"2026-08-15T19:24:58.272797Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.411653Z","title":"Imagenet classification with deep convolutional neural networks","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":68,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.411653Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:66033c8b321f81adca4f9eefaf09e78a061ab8f0586f51695ec794e68d1854e5","observation_id":"d584834c-e796-4a60-869f-a899c0391e67","resolution":{"observed_at":"2026-08-15T19:24:57.411653Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.415906Z","title":"Efficient backprop","venue":null,"work_id":null,"year":2002},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":69,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.415906Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:e81c8106e87052021738c270b6314c4f3efdcd145c52035b62814d2d6362e15e","observation_id":"98cf3d30-fa0c-4697-a8e8-06282af42d26","resolution":{"observed_at":"2026-08-15T19:24:57.415906Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.419621Z","title":"Deep neural networks as gaussian processes","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":70,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.419621Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:461393bc95a6871fda09c50124ab4fe9d1fc8bc512e6797446d1150f3e713b29","observation_id":"9e485507-1026-48fb-8827-8f99a8b9143a","resolution":{"observed_at":"2026-08-15T19:24:57.419621Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.01908","last_updated":"2024-06-06T09:07:37Z","snapshot_observed_at":"2026-08-16T13:46:28.797452Z","submitted_at":"2024-06-04T02:39:42Z","title":"PDHG-Unrolled Learning-to-Optimize Method for Large-Scale Linear Programming","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.01908","snapshot_observed_at":"2026-08-15T19:24:57.423732Z","title":"Pdhg-unrolled learning-to-optimize method for large-scale linear programming","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":71,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.423732Z"},"links":{"cited_paper":"/paper/2406.01908","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:4d356f612fb12f9a4a38cad7b3d93c7f449ed8664d740a692fe40f3ce268af51","observation_id":"97b0ae74-eba1-43c8-b21b-d8d2188dd870","resolution":{"observed_at":"2026-08-15T19:24:57.423732Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.427844Z","title":"Deepgcns: Can gcns go as deep as cnns? In Proceedings of the IEEE/CVF International Conference on Computer Vision, 2019","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":72,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.427844Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:5038f667995aafb97d16d74d1eb559f3af010af54f506f2942eb8b6f9949d699","observation_id":"f7d31d6d-eb62-4b9d-b16f-a2b5cdf8962a","resolution":{"observed_at":"2026-08-15T19:24:57.427844Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.431574Z","title":"Deepergcn: All you need to train deeper gcns, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":73,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.431574Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:535dbb20ad46f27affdb8afb11fe08a57babf956b756050f64059d44622f326e","observation_id":"ae96fa72-2d7c-4998-9627-6c52ce3475da","resolution":{"observed_at":"2026-08-15T19:24:57.431574Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.435829Z","title":"Training graph neural networks with 1000 layers","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":74,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.435829Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:2ae70ed4fd6e8399344108584aeba614b8590ef58732bf35313b2433d0f77c89","observation_id":"a5fdfd23-0b97-4ceb-bb15-4ad4704d2cfc","resolution":{"observed_at":"2026-08-15T19:24:57.435829Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.439693Z","title":"On the initialization of graph neural networks","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":75,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.439693Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:be07c0951c0b1c6366606fd0d3bcbb945ff2ef553083e41bc9969f42c413c67c","observation_id":"39d3aa15-7d9a-4622-83de-2c5806560381","resolution":{"observed_at":"2026-08-15T19:24:57.439693Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.443456Z","title":"On random deep weight-tied autoencoders: Exact asymptotic analysis, phase transitions, and implications to training","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":76,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.443456Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:6eacd216b8dfee8ccb5a722ed1fa0b416fd98f2fd58e6fb6ad9645dfef86a9f2","observation_id":"464cce68-f87e-40bb-b6e2-5d73449364d0","resolution":{"observed_at":"2026-08-15T19:24:57.443456Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.447304Z","title":"On the power of small-size graph neural networks for linear programming","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":77,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.447304Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:6c14c997136f87b6b7670d3aaa9329252cdef19a37fd13d2273653d3d5cfc38f","observation_id":"76dd95d7-44b1-4174-ba0a-cb8c4598a906","resolution":{"observed_at":"2026-08-15T19:24:57.447304Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.174624Z","title":"Deeper insights into graph convolutional networks for semi-supervised learning","venue":null,"work_id":"a90b88ff-def7-44fd-92ae-1aa62ff33a16","year":2018},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":78,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.450995Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:ca998c73aaaaf3897f5a54dbd3de30198cbfe3007f6c4a08d99fd59f0acd703d","observation_id":"05845693-62fe-47d8-92c7-ddde24a04d1d","resolution":{"observed_at":"2026-08-15T19:24:59.178707Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.161999Z","title":"Why deep neural networks for function approximation? In 5th International Conference on Learning Representations, ICLR 2017, 2017","venue":null,"work_id":"7e91c7d0-c88d-4c84-891a-316963a85add","year":2017},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":79,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.454790Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:85a2d8c9b4bf7f4794bffbdb302c3de485db0c4085b434dd1d652ce52e01bfe5","observation_id":"919b764a-b3c9-46e2-a4d2-087d0d23dbe8","resolution":{"observed_at":"2026-08-15T19:24:59.166244Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.148740Z","title":"Large scale learning on non-homophilous graphs: New benchmarks and strong simple methods","venue":null,"work_id":"b318de2d-22c9-40dd-b416-8b42782d237b","year":2021},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":80,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.459345Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:2b0c31330743759016d9fd8b4a13be17cc1fb1b0237c3fc155927f4559fe40f1","observation_id":"69bc9f17-2224-4cdf-ab73-5bc399db6164","resolution":{"observed_at":"2026-08-15T19:24:59.152768Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.135787Z","title":"Towards deeper graph neural networks","venue":null,"work_id":"5a12c602-5b6c-4f91-906b-9ff2acccd8bc","year":2020},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":81,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.464185Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:9b70812c8d4d4cf2a501390af19730dc993fb0be369f49903f3746d2855540f8","observation_id":"0aa74f5b-072e-432d-b9ee-2f58686f81d4","resolution":{"observed_at":"2026-08-15T19:24:59.140210Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2112.11628","last_updated":"2024-01-23T03:03:54Z","snapshot_observed_at":"2026-08-17T23:09:46.327006Z","submitted_at":"2021-12-22T02:18:31Z","title":"SkipNode: On Alleviating Performance Degradation for Deep Graph Convolutional Networks","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2112.11628","snapshot_observed_at":"2026-08-15T19:24:57.469947Z","title":"Skipnode: On alleviating over-smoothing for deep graph convolutional networks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":82,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.469947Z"},"links":{"cited_paper":"/paper/2112.11628","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:d410f77275a7ec67e8601ca98afab8a8ea46e0c58852b7c70f551b6202e39968","observation_id":"3c0f2f8f-95bf-4cfc-9315-b7a261e72da5","resolution":{"observed_at":"2026-08-15T19:24:57.469947Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.122548Z","title":"Break the ceiling: Stronger multi-scale deep graph convolutional networks","venue":null,"work_id":"df5d8082-c56f-470b-a467-43803a3d7dd0","year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":83,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.474318Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:906d0cef1ff5ce0c42199f3f9e9bbecf01b241229ea3ee87db6eedc0522f517b","observation_id":"55bff787-074d-4497-8d87-48a17e2c6dc2","resolution":{"observed_at":"2026-08-15T19:24:59.127095Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2008.08838","last_updated":"2023-11-03T22:30:21Z","snapshot_observed_at":"2026-08-14T20:40:10.175894Z","submitted_at":"2020-08-20T08:36:27Z","title":"Training Matters: Unlocking Potentials of Deeper Graph Convolutional Neural Networks","version":3},"cited_work":{"arxiv_id":"2008.08838","doi":null,"metadata_source":"pith","pith_arxiv_id":"2008.08838","snapshot_observed_at":"2026-08-15T19:24:58.221002Z","title":"Training Matters: Unlocking Potentials of Deeper Graph Convolutional Neural Networks","venue":"cs.LG","work_id":"35bda5da-6a19-43a9-add6-4307056c29fb","year":2020},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":84,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.478870Z"},"links":{"cited_paper":"/paper/2008.08838","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:d42d0149d4716647231d91872d7f56823421c3f5d72e4f79d93c685ea7ffeda8","observation_id":"97072f62-184b-4773-8dc4-803cd817ed81","resolution":{"observed_at":"2026-08-15T19:24:58.225419Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2406.08993","last_updated":"2024-10-28T09:03:11Z","snapshot_observed_at":"2026-08-18T09:26:29.486337Z","submitted_at":"2024-06-13T10:53:33Z","title":"Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.08993","snapshot_observed_at":"2026-08-15T19:24:57.483700Z","title":"Classic gnns are strong baselines: Reassessing gnns for node classification","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":85,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.483700Z"},"links":{"cited_paper":"/paper/2406.08993","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:3656312566ff2ae87434239cf0c144d73483bdf3a1351d93f50d6368e31c0572","observation_id":"ef6dcf14-ddb0-4c6a-8159-5e464006715c","resolution":{"observed_at":"2026-08-15T19:24:57.483700Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.109090Z","title":"A fractional graph laplacian approach to oversmoothing","venue":null,"work_id":"b8d719bd-928e-4694-bdf1-f28a24bed1f5","year":2024},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":86,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.487586Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:a55dbfb348dbafe216dcc7a134587dc8f380be0dc3cff9d1e6daf548a726530d","observation_id":"0e48f098-bc44-4680-bb26-853e79e69d74","resolution":{"observed_at":"2026-08-15T19:24:59.113803Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.095015Z","title":"Gaussian process behaviour in wide deep neural networks","venue":null,"work_id":"8fef1880-3b65-469d-b474-02c3a4b0c37d","year":2018},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":87,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.491329Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:712487a8f250510bd091253097b779cc1236744e49c75d0b1fa988eb27e0ee08","observation_id":"e773b079-42aa-4181-8859-1c0d318ec46a","resolution":{"observed_at":"2026-08-15T19:24:59.099939Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.080783Z","title":"Scattering gcn: Overcoming oversmoothness in graph convolutional networks","venue":null,"work_id":"91117010-459b-46f7-83e4-30d22ac2eb77","year":2020},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":88,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.496458Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:3494970182a063062869463210f6d6debb61058d9a2830f03c7a4e61ae53db89","observation_id":"3ba60682-76e3-47f0-b552-2ade60841c2d","resolution":{"observed_at":"2026-08-15T19:24:59.085480Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2012.13349","last_updated":"2021-07-29T13:41:25Z","snapshot_observed_at":"2026-08-16T18:56:28.610181Z","submitted_at":"2020-12-23T09:33:11Z","title":"Solving Mixed Integer Programs Using Neural Networks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2012.13349","snapshot_observed_at":"2026-08-15T19:24:57.500704Z","title":"Solving mixed integer programs using neural networks","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":89,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.500704Z"},"links":{"cited_paper":"/paper/2012.13349","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:eb4c2f68d6f2058795c0127718316e24acbea1a09e37ebd05ac5fc9026700957","observation_id":"0861386b-bead-4f09-927c-a1a06e218e10","resolution":{"observed_at":"2026-08-15T19:24:57.500704Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.065446Z","title":"Bayesian learning for neural networks, volume 118","venue":null,"work_id":"9fc14e21-4648-4b25-bc4d-6f47f2b38a3b","year":1996},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":90,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.505484Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:b26541638f1b460503cd743bf3f76ba8dab1ee776a5caa3c0e127efaf8985f41","observation_id":"0d56b641-d36f-4fd3-bbb5-783773e5d91a","resolution":{"observed_at":"2026-08-15T19:24:59.070836Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.510675Z","title":"Random gradient-free minimization of convex functions","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":91,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.510675Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:00c93046cd712af108aea55a1dea21cad23d88b6edf8617ca1eec8676f4ebdb7","observation_id":"0ffdf19a-14d5-45ae-bb38-b5a372448059","resolution":{"observed_at":"2026-08-15T19:24:57.510675Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:57.516955Z","title":"Revisiting over-smoothing and over-squashing using ollivier-ricci curvature","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":92,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.516955Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:937fae1528315d176390c2a0d7275bd9839930da5b80a3c40ef1fe4bcecb633a","observation_id":"d1029184-dd79-4dfc-9fbf-c5cb31389c97","resolution":{"observed_at":"2026-08-15T19:24:57.516955Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.035739Z","title":"Graph neural networks exponentially lose expressive power for node classification","venue":null,"work_id":"778bfe8e-a313-4ee2-beac-e6993df051af","year":2019},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":93,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.521664Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:b891a7959d84334d8429bc7cef53396a74135af66c2d0173555df5970df767bb","observation_id":"d1a39ea9-c899-495c-b66e-7e8fb88b1524","resolution":{"observed_at":"2026-08-15T19:24:59.040403Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.022666Z","title":"Applications of combinatorial optimization","venue":null,"work_id":"80e7bb47-59a6-434d-835c-fcee384d0bcb","year":2014},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":94,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.527478Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:27cb30e02852c60a6d0f8780093ae3373183596b52e6ba730094b3caaad6eccd","observation_id":"2a243edb-a5e1-4e6d-b4ef-4aa14094aac5","resolution":{"observed_at":"2026-08-15T19:24:59.026991Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:59.008324Z","title":"Resurrecting the sigmoid in deep learning through dynamical isometry: Theory and practice","venue":null,"work_id":"8d82d04b-1e8e-43b2-a364-2841878cfc46","year":2017},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":95,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.531659Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:a18c60841470430441c4edeb8a68f3785d5937056fd2f7c52988c0a087a705d7","observation_id":"acf57eee-1117-4469-8360-e89e9ab3e866","resolution":{"observed_at":"2026-08-15T19:24:59.013044Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:58.993064Z","title":"The emergence of spectral universality in deep networks","venue":null,"work_id":"4843c57f-b519-4b40-9981-6d42d7ac42e7","year":1924},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":96,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.536485Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:a32d2b8da96b547f445042f1d5affb0f7bf3d4d7bd23f3192826e86db7e6b6b8","observation_id":"caece68e-d163-45a7-b450-4279ce076e23","resolution":{"observed_at":"2026-08-15T19:24:58.998997Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2302.11640","last_updated":"2024-03-02T21:17:13Z","snapshot_observed_at":"2026-08-16T15:53:25.866662Z","submitted_at":"2023-02-22T20:32:59Z","title":"A critical look at the evaluation of GNNs under heterophily: Are we really making progress?","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2302.11640","snapshot_observed_at":"2026-08-15T19:24:57.540519Z","title":"A critical look at the evaluation of gnns under heterophily: Are we really making progress? arXiv preprint arXiv:2302.11640, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":97,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.540519Z"},"links":{"cited_paper":"/paper/2302.11640","citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:69cdfe1913c21c173fa64661abc4ff8f02d51215458c8e34c96d2663b4355453","observation_id":"79abcf6a-45fc-494f-8f69-f65c71132e3a","resolution":{"observed_at":"2026-08-15T19:24:57.540519Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:58.975607Z","title":"Exponential expressivity in deep neural networks through transient chaos","venue":null,"work_id":"9e537a2c-4cb6-45bc-8e04-2715a412771a","year":2016},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":98,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.545448Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:b80e87b4c47888e002b4f1fb631d8d96a4ee3e7f229728dffb065ca606ada53c","observation_id":"64916308-d86e-4cb9-9ca8-94efebc62be7","resolution":{"observed_at":"2026-08-15T19:24:58.983544Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:58.956849Z","title":"Exploring the power of graph neural networks in solving linear optimization problems","venue":null,"work_id":"ad04c555-6624-4da6-aebc-858f70e4b80f","year":2024},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":99,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.549735Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:ff9ef86ae5f033267af45c79959482a20529429bb88fbc3afd199e6a851c6c9d","observation_id":"8c0266ee-4fec-4b6c-8afa-2482d1c8af11","resolution":{"observed_at":"2026-08-15T19:24:58.961682Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T19:24:58.942530Z","title":"Dropedge: Towards deep graph convolutional networks on node classification","venue":null,"work_id":"e5c6f128-6e15-40ca-a8ce-df45f8e0e7b8","year":2020},"citing_paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective","version":2},"reference_index":100,"source":"arxiv_source","source_observed_at":"2026-08-15T19:24:57.553636Z"},"links":{"citing_paper":"/paper/2506.16790"},"observation_digest":"sha256:66e72df8d86cfa700bb0a5479ab4865bfb98fd4dbf21f06ccf6f2c21f5e84296","observation_id":"3f34df87-c140-425a-b4e8-f5be47f3f8a6","resolution":{"observed_at":"2026-08-15T19:24:58.947191Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2506.16790","last_updated":"2025-07-15T05:21:45Z","latest_version":2,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-18T09:27:08.728387Z","submitted_at":"2025-06-20T07:14:31Z","title":"Exploring and Improving Initialization for Deep Graph Neural Networks: A Signal Propagation Perspective"},"reference_resolution":{"displayed":100,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":80,"verified_exact":4,"verified_fuzzy":16},"total_outbound_references":147},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"thesis":"As of 19 August 2026, this Paper Citation Record lists 100 of 147 outbound references and 0 inbound Pith citation observations for arXiv:2506.16790."}