{"as_of":"2026-08-12T20:13:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:068c9dd9d1b6aa571b0e9948098271e16889be9ee09c7a3d304a6bcf8718451f","coverage":[{"denominator":119,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":100,"source":"paper_references, paper_reference_links","source_observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-08-12T06:34:41.77262+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/2607.03784/citation-record","integrity":"/paper/2607.03784/integrity","json":"/paper/2607.03784/citation-record.json","paper":"/paper/2607.03784"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"Token merging: Your ViT but faster","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:648e832645dcf618abc592e3f031f32cd96d998cb75770cd5d8d7661c5017fe6","observation_id":"0ae8267c-97c4-4c47-8491-e93348d3b5a1","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:bbb89cb3d6756707fe17564112cd24a10a93e222d73d920a23f42e0e8f012634","observation_id":"07c0d54a-26c7-4566-bbdf-aa9d7a3503ff","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Transformer interpretability beyond attention visualization","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:a972b5f99711909b83c6039cd73c86365d1b30b2b61d549192629dfe33420868","observation_id":"b0bd2002-fc72-4a33-89ed-d9325a67835f","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"R., Fan, Q., and Panda, R","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:ac2937934f55d4510e74db0025740856e9447c74f7be67721a4a2ffee1fc05ce","observation_id":"847015b1-ef50-4183-a865-1ba1a5f393c4","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Diffrate: Differentiable compression rate for efficient vision transformers","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:a3ebcce1b95b6b065669aa2806dbc8b1c8a972e7403df7fc5994cb0646c367b8","observation_id":"d7a0b12c-eee4-4d7d-a328-a6a5d86ad0a8","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Chasing sparsity in vision transformers: An end-to-end exploration, 2021 b","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:d984a0a934a22e6cff83882803409af9028eb7a40ccd5b06d88c43803989c152","observation_id":"4352ef5c-6c02-4dc3-879c-4ac3f437794c","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"SparseViT : Revisiting activation sparsity for efficient high-resolution vision transformer","venue":null,"work_id":null,"year":2061},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:ecd11b82583348627dd2f5865538396e54106b292d866bb3548eb92c5960569d","observation_id":"30291fc2-36b1-4d12-8efb-8fa472e706b7","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"The stone-weierstrass theorem","venue":null,"work_id":null,"year":1959},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:47b7d0dd6630489d73f05bc4e3b2f8c100d0fc17195ad22fb2050327b50a818a","observation_id":"1f8c303b-fe1f-47d0-aaec-51855a7a42bd","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Imagenet: A large-scale hierarchical image database","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:e396642b3a4c8eb55c9b32399b41e8e290ad7a030e9f51fbf3dc0375111f5772","observation_id":"e1b042ed-7444-4ad1-b546-4ee3880fe682","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"An image is worth 16x16 words: Transformers for image recognition at scale","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:077c7157459bd12c1f4f69cbe911cee54337c0f9120f93ce5d5c51b1f29d6bcf","observation_id":"051bbb16-b76b-439f-a06b-8d90676003d6","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"and Chong, E","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:5eda59139f67d70c763b38fff18f1832a0e2caa16da8ef0f5817d63384d54066","observation_id":"64fe9407-2350-41b5-ac2b-58d81f01bb62","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Transformer in transformer","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:11bbcef172b887349dec91212c18a62a130bfcdaf26258f0b6071c3b734d59ff","observation_id":"cfd5e04b-1ea7-4b0c-b8f1-3d173601849f","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"and Xiao, L","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:f1bacf4169aac71ca2b528a4efd9255a5b903fc3d5e7a53e27305e892383d433","observation_id":"1088c285-c0f6-4a3c-a30e-36efb76bfa22","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Learning multiple layers of features from tiny images","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:0561f0207f742feb0cf01a4627da9fa8e2192e543779f8f69bf29b10013d5575","observation_id":"25f617eb-4cb6-4cd8-87f7-da9d1dff8696","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"EfficientFormer : Vision transformers at mobilenet speed","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:c95711d631be028da4d5674f12812720a0107b5ee91c1a291d8d572b06ae7181","observation_id":"63c2d926-17f3-4af8-a48f-cdb5ec8fb247","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2202.07800","last_updated":"2022-04-13T23:46:08Z","snapshot_observed_at":"2026-08-03T03:15:29.208149Z","submitted_at":"2022-02-16T00:19:42Z","title":"Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2202.07800","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"Not all patches are what you need: Expediting vision transformers via token reorganizations, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2202.07800","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:3262e68b95b61bfe45992a9fdf29170fdac72a5b9fb81304a99c31cd0a523cff","observation_id":"4df548e4-465b-4e2d-ae53-08007170b7ce","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Mlp can be a good transformer learner","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:d23753dc4d75f8c5a1cd2116a399d06383f0541ce44792fba616b2cd5418a683","observation_id":"92dd261d-1a61-4036-99c4-b7b0d07e61ac","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1405.0312","last_updated":"2015-02-21T01:48:49Z","snapshot_observed_at":"2026-07-06T03:42:37.507458Z","submitted_at":"2014-05-01T21:43:32Z","title":"Microsoft COCO: Common Objects in Context","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1405.0312","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"J., Bourdev, L","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/1405.0312","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:7539f18eaeda2d7c8381798c802069fb664964de45be20189f10ab07cfc3e5e3","observation_id":"6a22cba9-7f99-4a0d-ae6a-f16363f321a0","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"TB-STC : Transposable block-wise N:M structured sparse tensor core","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:3a31e8aa19df97812f01f628630f8c1b23258973fa598f672e1ecc6f088a47b6","observation_id":"a6c27d2a-1c33-406a-926d-3c424617fe78","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Swin transformer: Hierarchical vision transformer using shifted windows","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:0bea0998cdef393f39cf454e9c5d0b7249b59e370afa47ff550f9cf010c19fd9","observation_id":"f14bdd58-cf48-47bd-a1f9-f43ea63ad448","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:ecf49402d0f4ff8f0939d5bca964da4b3d39c0be54215858b75cb75822a0dab1","observation_id":"b44af04e-b7e0-437e-bf9e-31bfc013a158","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"ShortGPT : Layers in large language models are more redundant than you expect","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:97abb40b1eb911eb5f69df932bb1009fe22d0fe5183ce27716034fcbbd634bb1","observation_id":"b8062bcb-5604-43c5-a09a-d12e9558a6c8","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Speedup deep learning models on gpu by taking advantage of efficient unstructured pruning and bit-width reduction","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:2e1ba53911e24691bd5b4315d35d7896af5ac102413a2f0ee6cedb7d90acc234","observation_id":"e2aa63fb-c74f-44b7-a743-814b14f96642","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"DynamicViT : Efficient vision transformers with dynamic token sparsification","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:bac74c03ac2fadac64f858b2e117b1795802af7a7d34797308a7d6243f31b243","observation_id":"01806e45-8eeb-4007-a094-11ffbbf1ff05","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2306.11695","last_updated":"2024-05-06T17:47:01Z","snapshot_observed_at":"2026-07-06T15:44:42.776459Z","submitted_at":"2023-06-20T17:18:20Z","title":"A Simple and Effective Pruning Approach for Large Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2306.11695","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2306.11695","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:6d4c5017a75d6e20b878a1404d298ba40a8406d9b9bd449afe73fb08b289ecbd","observation_id":"ac72c40c-311c-4ee3-a5ea-85398b211105","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Going deeper with image transformers","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:dbcac474d99ada3cab06812d8e5ebb25824a3421c9f0d75442032b458396665c","observation_id":"a43519db-ed42-408d-84f5-0a1ec4790d20","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Regression analysis","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:512902bf7af1bccf21ab1a5da303cc0b935c91ece1db0ecae6e49931d915c19e","observation_id":"7b1e520f-54a7-4e0d-940e-bdf30fd3ed01","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"PVT v2 : Improved baselines with pyramid vision transformer","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:89576aeea1c419f7c15366cbe99e55a89d49f56051600a48914d0b552e19632a","observation_id":"04b86278-4084-4b15-a3aa-c7d48d94f54d","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Joint token pruning and squeezing towards more aggressive compression of vision transformers","venue":null,"work_id":null,"year":2092},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:728692bef6e1d64a41f14efbf1782dedbf2d345739ebdaa060022da63b66cfc1","observation_id":"df03824a-4076-42da-931b-6deff9e67332","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"GTP-ViT : Efficient vision transformers via graph-based token propagation","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:1550520e92e7c48c35c26ff9aadfd44fea29568edddd4519b352aed44993cb91","observation_id":"d5faabc3-1126-4dfe-a9cd-30c2f8d80bbd","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Evo-ViT : Slow-fast token evolution for dynamic vision transformer","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:9b545e4a81466f95bb47fb53617bb9f31027bf172cc4c02c2c8c82867d904186","observation_id":"9fbebde7-5d71-4a5f-8261-18cc0f7bce9e","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Global vision transformer pruning with hessian-aware saliency","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:231dbf0dd79590c8abfcf4fdfd603d213b975a4e76ba250bc4826392096ee0ee","observation_id":"db088654-d4a2-423a-8386-1a05ea28bea3","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"10.1609/aaai.v36i3.20222","metadata_source":"openalex","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Width & depth pruning for vision transformers","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","work_id":"99f38098-5a9e-499f-a024-79ca1addb529","year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":42,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:fb58e211521038de4e9e98a7ef50a96ac8aaccdeab5660fb646a5389cc53b7f2","observation_id":"1b04340b-13e1-490d-9e88-c100b4b66407","resolution":{"observed_at":"2026-07-12T00:08:22.178374Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-12T06:34:41.77262+00:00","source":"crossref"},{"observed_at":"2026-07-12T19:49:18.907403+00:00","source":"crossref_status_cache"},{"observed_at":"2026-07-12T19:49:18.907403+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-12T06:34:36.333875+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":"2026-07-11T23:58:47.097757Z","title":"Unified visual transformer compression","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":43,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:b712ce3adf83db4d2935b319054db7920d65adfe5939209c730cc209e73871fd","observation_id":"a5b07959-e09d-47ee-ae20-99210bd4a2bb","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"SAViT : Structure-aware vision transformer pruning via collaborative optimization","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":45,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:e402f32b03c64299e73b040a6edc2048fc728a5c0e309fac34cc681b886f04bf","observation_id":"0d51d5fd-6f5c-4271-b685-8095b204c372","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Scene parsing through ade20k dataset","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:d5454b9def27b4bd27968c7f59a9607012f4195a294beaa20ae1f10967a42aec","observation_id":"1ba1fd19-0013-4581-af90-771624c5cb29","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2104.08500","last_updated":"2021-08-14T06:06:37Z","snapshot_observed_at":"2026-08-08T10:50:18.507454Z","submitted_at":"2021-04-17T09:49:24Z","title":"Vision Transformer Pruning","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2104.08500","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"Vision transformer pruning, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":47,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2104.08500","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:00c45963ce13b33656cc072cdf380bc0ae8b9dba73b0ab0ef88adad58820e3b7","observation_id":"40d320c8-9551-4519-82da-f040833a76f2","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"IEEE Transactions on Pattern Analysis and Machine Intelligence , volume=","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":48,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:1ea1598a699125528ea0518d4258f3db0ce6d4204e96ef7d827801456fc6a6cd","observation_id":"23eb139a-1fb7-4fc1-87ad-df8731685b3b","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops , pages=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":49,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:8096413aeab6f7484aae57f465ef38e620aaa9bd22b64022c09645a32cb33cbc","observation_id":"e7b5e6d4-7cca-4c6c-8c3b-780a2ad7839f","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.12837","last_updated":"2024-07-08T04:55:34Z","snapshot_observed_at":"2026-08-10T14:08:05.183675Z","submitted_at":"2024-06-18T17:55:15Z","title":"LayerMerge: Neural Network Depth Compression through Layer Pruning and Merging","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.12837","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"arXiv preprint arXiv:2406.12837 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":50,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2406.12837","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:7a6a981ea87cd1a727add118de7421e762d5fbf0ae88ffa1a85f529f1df8de5b","observation_id":"3b6b2445-471d-4ec0-b307-2f0e8d18600c","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"International Conference on Machine Learning , pages=","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":51,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:497759860831e9f4ba813aa591b3256a300932130fe81d00a88be10dc9661df1","observation_id":"55236f18-2388-4950-a1de-b14b59c16621","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2106.09309","last_updated":"2021-09-03T13:02:30Z","snapshot_observed_at":"2026-07-06T11:20:12.897633Z","submitted_at":"2021-06-17T08:22:46Z","title":"Layer Folding: Neural Network Depth Reduction using Activation Linearization","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2106.09309","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"arXiv preprint arXiv:2106.09309 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":52,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2106.09309","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:1f533bcf105522e7a25605bf1514389866ecc4a43510cb41f88b85e2d6a89c16","observation_id":"511ae96d-cd05-4efc-9cca-ffbc0fc106a8","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Langley , title=","venue":null,"work_id":null,"year":2000},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":53,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:8ebbd20c485484c878cd22702f993aa2a1af71de588a9029602679a36f863cff","observation_id":"08e5474b-b12f-4574-98fd-26dea78f23a5","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":1980},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":54,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:f920df673f2c5c3853eef9ab25737893155c6c3372a665786eede69e930fce23","observation_id":"5cbecbf0-1a82-4a9c-9cfd-01dbbdf25286","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":55,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:bda823e65f429125c35e8c1970c6e725dcfed807ed40beaef14b78b2a081c60c","observation_id":"6af8121c-4791-4124-b587-452a658b2070","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"parse_uncertain"},"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":"2026-07-11T23:58:47.097757Z","title":"I , publisher=","venue":null,"work_id":null,"year":1983},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":56,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:0b6d66bc475da93708139e2c4b83645666a637ebf3b21c98a73ec2437ae55802","observation_id":"57728754-ff6b-4ee0-8379-3e497b009c5b","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":57,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:7edaa1a6823645e65fbde013520bd314e6223908d3d60606ccc10856ec9fe7c1","observation_id":"d45feb57-6bcd-497e-994d-5059ecf87634","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":58,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:0e52a2c57ce5f7aa6402ce9caa25419e3056fc29cc682ce0d8d62eb47dd86e53","observation_id":"026c97ce-78d5-4527-993c-5fa3bba66d8d","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Newell and P","venue":null,"work_id":null,"year":1981},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":59,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:bcd46bbdb5681a273eea97b9b1357680d497377a4bd78999d48ea97a4710b3f2","observation_id":"c6129384-6388-46c1-9ffa-d845309f3a62","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":60,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:3fe9c859e4de0a748c2bbb929b782840e5f75d2f52bbd8531320f50084e0af90","observation_id":"b2e160fc-5da8-454f-bf0b-35bee85c1ab0","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"and Kaiser,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":61,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:e9088f2fff3c3cf73fd8b65da82a44c9b69134fcc063e8bc6225ca21b6df2d5f","observation_id":"5b84ea92-2a9d-4875-8984-b080b87b9d8e","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"International Conference on Learning Representations , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":62,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:fda27a0214e5ee61e9fcc953de94dd25836b6ed09cc235977920e3b500ad3e8a","observation_id":"885314ea-79cc-447b-896e-a46e54c0cb37","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"International Conference on Machine Learning , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":63,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:24ce2077e69b431deeea713c812eae1d2505c9049db98efed1b89b58abdee374","observation_id":"5addb896-c098-444a-a6a8-db7ca76139c3","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction Without Convolutions , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":64,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:75a0622473db584dfc08c4cb5c7462e7c7a0864ba523cc7217ecaf111507dfc6","observation_id":"463ead81-458b-4ec9-88a6-18e6b547f59e","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":65,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:fbc9eaa207869030a95243581566d95f8ce1c5ac768acfc45bcd546488255939","observation_id":"c0ec6fce-419b-46a8-bb87-e6a560890687","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2112.01526","last_updated":"2022-03-30T17:56:37Z","snapshot_observed_at":"2026-08-10T16:05:12.638929Z","submitted_at":"2021-12-02T18:59:57Z","title":"MViTv2: Improved Multiscale Vision Transformers for Classification and Detection","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2112.01526","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"2112.01526 , archivePrefix=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":66,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2112.01526","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:28468f2798b03466d079a32426f314a27d2c6355bc2e5f5637b085cd92552b14","observation_id":"6dfce47d-8805-492e-b407-0cba9d26e496","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"and Tan, Mingxing , booktitle=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":67,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:fa4976d900a599165a9abd769a0cf8e2edeeb509ce02a30bb855834935532931","observation_id":"bf0059e4-f889-4257-bd8b-5326b2da552a","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2021 , eprint=","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":68,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:f3cbccfd3e8334f0bc27b554e81a8b4f73486c3dec5f2550a1321ed6b05d3799","observation_id":"d2d03b30-ccd9-4555-a1cd-9f843b50a1c3","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2022 , eprint=","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":69,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:828696e2d6d9057ae2c231df46c307ea70a2f61e099d38dfcc85c632cd4e65c3","observation_id":"a371dac9-2e00-4584-ae8d-0b06915b45cc","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2107.00652","last_updated":"2022-01-09T05:49:30Z","snapshot_observed_at":"2026-08-04T23:06:17.448062Z","submitted_at":"2021-07-01T17:59:56Z","title":"CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windows","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2107.00652","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"2107.00652 , archivePrefix=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":70,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2107.00652","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:35fd45aed22a426739a70106d9b8a1e1ffde62f832f48c733079ddff888cc26d","observation_id":"f6870304-4209-4791-abb8-e0e53b3d8fbe","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2204.03645","last_updated":"2022-04-07T17:59:32Z","snapshot_observed_at":"2026-07-06T12:57:58.716731Z","submitted_at":"2022-04-07T17:59:32Z","title":"DaViT: Dual Attention Vision Transformers","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2204.03645","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"2204.03645 , archivePrefix=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":71,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2204.03645","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:f7362ac7a24c991517953fa3cb86d3fadeec83f9b05079b9ee0ec8e6b7cd0791","observation_id":"8bb674fe-7cc5-441b-bef4-8809a149bbc2","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2203.16768","last_updated":"2022-03-31T02:55:39Z","snapshot_observed_at":"2026-08-11T14:13:02.914442Z","submitted_at":"2022-03-31T02:55:39Z","title":"ReSTR: Convolution-free Referring Image Segmentation Using Transformers","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2203.16768","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"2203.16768 , archivePrefix=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":72,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2203.16768","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:1cbb3223daa3e9ce5b187f4a8ca832ddf6f18e9e2400e1388052ae275b7377df","observation_id":"f5109bef-e6dc-4659-919b-37868cdb4df2","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2022 , eprint=","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":73,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:5626eddbd7ea253bfc9f64a8f6c35558deeec1e3f772613d809ad4411a781e23","observation_id":"f5089e8f-5250-468f-9f5b-f6a018614df1","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":", title=","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":74,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:3aba87086accced180803894782728a51aba8847faa0299fab058ef4904d1cec","observation_id":"f816eee1-82aa-4ba3-ab17-c46d26baf3c9","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2021 , eprint=","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":75,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:0d9dd5f7ef0f291a3ef4f34df07507f523eaec6e40c138c25fbc209546950a5e","observation_id":"a82b29bc-7a1a-4fce-9860-6a3cd841f8c2","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"International Conference on Learning Representations , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":76,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:fb2790bc03c350ad5aec49829abdad6a0852789baf32da258e750a66a81555cd","observation_id":"ad94c89f-ee55-4235-b5a2-306175190648","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":77,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:1594d93d75d6193a104e5e4adcb45d7714c9df846afe26e12d4b11ceb2828482","observation_id":"5af6fde3-221a-4521-9f7d-08f8f8ce68b3","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"MLP Can Be a Good Transformer Learner , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":78,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:d283504c5291820ab51ea5819dfc6de95d7ae975368bd4e328a866bf3bf1bfe4","observation_id":"0aaf70c9-2b41-41ee-80c6-48fa01ce6e48","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2206.00843","last_updated":"2025-01-04T03:32:03Z","snapshot_observed_at":"2026-07-06T13:16:43.219735Z","submitted_at":"2022-06-02T02:32:47Z","title":"DepthShrinker: A New Compression Paradigm Towards Boosting Real-Hardware Efficiency of Compact Neural Networks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2206.00843","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"arXiv preprint arXiv:2206.00843 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":79,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2206.00843","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:267bde5f05524bcea93bbefe1f668d3d435ff723e501b6aaadd64f2e96bd3f18","observation_id":"ca95a55c-e266-4baf-ab0b-31192ebe7291","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2021 , eprint=","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":80,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:f4add59c09ef78996c27087412ff06a779f1b2c80de6dadbdc2ca361e21e406b","observation_id":"0d9f6259-4b42-4495-ac5f-7b0a07be2f88","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2301.00774","last_updated":"2023-03-22T12:33:46Z","snapshot_observed_at":"2026-08-12T15:19:45.497046Z","submitted_at":"2023-01-02T17:48:56Z","title":"SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2301.00774","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"2301.00774 , archivePrefix=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":81,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2301.00774","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:1b9e0761fb721e132a44cc57b6230596c35b352256881fe052a8d8f38b1c8b09","observation_id":"b7223d71-b82a-4697-ad7c-02e37c04eff3","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2024 , eprint=","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":82,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:20cf8023e7c90d1323f9f9f13425f516a6cf3dc3abe0396436b07388ecf633d1","observation_id":"736184c1-5c12-49a6-893e-bad7cbc93d21","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2302.01172","last_updated":"2023-02-02T15:49:03Z","snapshot_observed_at":"2026-08-10T07:04:30.020413Z","submitted_at":"2023-02-02T15:49:03Z","title":"STEP: Learning N:M Structured Sparsity Masks from Scratch with Precondition","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2302.01172","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"2302.01172 , archivePrefix=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":83,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2302.01172","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:aff0d4f6972710725000c1d3f446e44e3c584662599b7b7a1e549e4579a22b6d","observation_id":"8ae2733e-b564-4906-a101-b519c7b9b538","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.02068","last_updated":"2025-04-15T07:54:24Z","snapshot_observed_at":"2026-08-11T12:42:50.994973Z","submitted_at":"2024-07-02T08:58:19Z","title":"LPViT: Low-Power Semi-structured Pruning for Vision Transformers","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.02068","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"Sabry and Yang, Xulei and Wu, Min and Li, Xiaoli and Lin, Weisi , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":84,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2407.02068","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:56a1e9bba1fa9a28e2c2599ae50fa97bd03ec1613dbaf722a3ebddd9964891db","observation_id":"13c9a042-2ecf-44db-baa1-44ba71f226c9","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.16135","last_updated":"2025-06-03T03:31:41Z","snapshot_observed_at":"2026-07-06T19:37:11.636987Z","submitted_at":"2024-10-21T16:00:04Z","title":"Beyond 2:4: exploring V:N:M sparsity for efficient transformer inference on GPUs","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.16135","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"Beyond 2:4: Exploring","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":85,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2410.16135","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:d331435cf96dfc0bb5f26444ea5eaaaa61f9042148baddba7dfe32973dbf1875","observation_id":"a9432fbc-1918-4d28-9e57-20a9e1c6e2e0","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":86,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:0bdfb65c9995dcd648b8474a1a27277908cd5781492ff991c2df2efa66ae9900","observation_id":"844dc753-87d8-4b35-a01a-38b8ee0b0294","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2022 , eprint=","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":87,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:66562be83a826d4cbca3ca751f7153cfa658969233ca8e4c9541a57e7144c652","observation_id":"9df25f5a-faaa-4434-95a9-7fbf25259ca9","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":88,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:0fa75bf0ebd0a9ef042a2841d158b64bc6d0a3223d25f51eada05139f8939e65","observation_id":"4abdf15c-d573-4ff4-8c1c-7c53de180eb6","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Proceedings of the 36th International Conference on Machine Learning , series=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":89,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:fda9a6790c501df4f69bcf996f55d4b4e03d0ecf79c5ceb5bca0d0fd31002039","observation_id":"64ee8947-f5e7-4c77-a357-9f8e448481c2","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2019 , eprint=","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":90,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:e2ece233f7ffea69117da88820541048a0c66d609b8390c5e4026fc05d603c0c","observation_id":"7d9f687b-00ba-4f6d-9f1a-bdf305cda8e5","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Proceedings of the 36th International Conference on Neural Information Processing Systems , articleno=","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":91,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:298f889f89cf6e8e5902063f11f6404ea28e632bb8b318b37bd10a4381a7cb9d","observation_id":"b0965a8e-140d-46eb-920e-5a864285f65a","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"ImageNet: A Large-Scale Hierarchical Image Database , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":92,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:0538a0e36ca26957d0e9ca51593c31b0937c5236d370651b8e223173da7fc437","observation_id":"b49786ec-a482-4110-8ecf-5d75f205b201","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Scene Parsing Through ADE20K Dataset , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":93,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:cd59e7d9e96c67d411ba362ff62d47490baefbba70d793bf7f25e49d96fb0e3e","observation_id":"14523b54-f891-4901-be61-c72039ce40dc","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2301.01296","last_updated":"2023-01-03T18:59:54Z","snapshot_observed_at":"2026-08-12T19:14:51.908052Z","submitted_at":"2023-01-03T18:59:54Z","title":"TinyMIM: An Empirical Study of Distilling MIM Pre-trained Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2301.01296","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"2301.01296 , archivePrefix=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":94,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2301.01296","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:10d394567d9eff90aaa5252f21cb2ab0d46615fd079f76487ec075dae3ed0b74","observation_id":"8afe9647-2096-4543-9b32-32dfe02af5ef","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Token Merging: Your","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":95,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:396d0622cc57e1c741e25329f9e996dd3bb9fa0d1f61b749281b72ddf6c80e89","observation_id":"c3a68de2-506d-47a8-a342-2c3a810511f3","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"International Conference on Machine Learning , pages=","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":96,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:e2a6da8f736f52efe6c2ba2350fd8143e7ccdd97ada4039cee60fbf4c9a03f5c","observation_id":"692a2a5b-e575-4785-8da5-8baea642c5d8","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1503.02531","last_updated":"2015-03-09T15:44:49Z","snapshot_observed_at":"2026-07-06T04:11:24.157003Z","submitted_at":"2015-03-09T15:44:49Z","title":"Distilling the Knowledge in a Neural Network","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1503.02531","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"arXiv preprint arXiv:1503.02531 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":97,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/1503.02531","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:fc4951dd839fe87ed38a0afd29180bb1ebb73fb3f7ac2cf76bcf4600e887854d","observation_id":"00420882-5931-4c09-a230-a4c3da727910","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":98,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:558588ea14649d381f89efea090a03a43268c53fc0d789cdfd46b939fa6017fc","observation_id":"c2f4c1ad-5a9a-42bb-9753-a8cab0487ae1","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":99,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:1bf2cb5028bf963599020e4ee3f8ed9da7fe3c7486165d0968ddf2f4358b5fc9","observation_id":"f0d26030-461e-4f2b-896d-efcd834df048","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Proceedings of the IEEE/CVF International Conference on Computer Vision , pages=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":100,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:da01dc3f32607c06dc502f5be734ad8c27a517163d37ff1c8c70285a7b7fa348","observation_id":"af91a26b-74e7-43f4-8a2f-e4bb327ad8b8","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , month=","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":101,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:75e4158c9842ef424d3f9921ddb87d140f9b5f08074bf18cafafcbdd149787a2","observation_id":"1740f731-5b07-4ad0-9290-aa14e2191270","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2022 , pages=","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":102,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:34d0727336fafae57bc3918a05eedb0d43c496146fcb39ee3abd3a7697698037","observation_id":"fe0a32fc-550b-4399-8ecc-d1f2ebf1a602","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":103,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:74c7037da627a19902b23110f0c6c0b4011252e0a5c76ebd88be368ace937c78","observation_id":"838b1525-fc9f-4e7a-aaeb-1b0b6395057e","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.01812","last_updated":"2024-02-05T09:21:28Z","snapshot_observed_at":"2026-08-10T17:12:13.544955Z","submitted_at":"2023-10-03T05:55:11Z","title":"PPT: Token Pruning and Pooling for Efficient Vision Transformers","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.01812","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"2310.01812 , archivePrefix=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":104,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2310.01812","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:687029a15b52648b4e42ab9d6a0133b1a8bed82c1cbf27da54bbb7f6ca789fa5","observation_id":"a2cceec5-58a5-4bc2-95ff-ce047d6a1a9b","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":105,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:e2d3ed5fcf4459b726db82e20cb7d9b54141878b690154718191b4ba8e9da053","observation_id":"2904b21a-f889-4a5b-bdd2-3b66c48da0a4","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"Proceedings of the AAAI Conference on Artificial Intelligence , author=","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":106,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:04454e3ed8d9c63614b3527a39caac191be3c856c2829aa8be48d8d90965680a","observation_id":"aa276d23-2715-45ca-b524-ab78378f6ff0","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.04616","last_updated":"2024-07-05T16:14:53Z","snapshot_observed_at":"2026-08-06T03:12:23.611292Z","submitted_at":"2024-07-05T16:14:53Z","title":"Isomorphic Pruning for Vision Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.04616","snapshot_observed_at":"2026-07-11T23:58:47.097757Z","title":"arXiv preprint arXiv:2407.04616 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":107,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"cited_paper":"/paper/2407.04616","citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:fabbe0b6259a807b1fb49d1a3c99e845a5d847d0d008d6808e5e5161f6281a7d","observation_id":"d6eda98a-6796-4739-8435-6c219d69757a","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2025 , organization=","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":108,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:2a8db8bd970753e624afc584c768d63eac3ee1ed6db5a59c2506ce78790313ac","observation_id":"ea73e555-4758-4a7b-860f-2805a974ec2c","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":109,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:e3a1fdb390d61857ed22aea2a26670ee6441e90ebb366255d7ce09f63123e1e5","observation_id":"78c485b5-d50c-4106-bbe0-3a6df73ca18a","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","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":"2026-07-11T23:58:47.097757Z","title":"2023 , organization=","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective","version":1},"reference_index":110,"source":"arxiv_source","source_observed_at":"2026-07-11T23:58:47.097757Z"},"links":{"citing_paper":"/paper/2607.03784"},"observation_digest":"sha256:16bd261502107ab5b6aa5756c2c15f6ef11dbfa40f0ee338a727d5075a670528","observation_id":"6e78e63b-c186-4921-9791-65c1aac210af","resolution":{"observed_at":"2026-07-11T23:58:47.097757Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2607.03784","last_updated":"2026-07-04T09:17:25Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-06T21:45:38.793670Z","submitted_at":"2026-07-04T09:17:25Z","title":"Rethinking Depth Pruning for Vision Transformers: A Heterogeneity-Aware Perspective"},"reference_resolution":{"displayed":100,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":1,"unresolved":98,"verified_exact":1,"verified_fuzzy":0},"total_outbound_references":119},"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":"2026-08-12T06:34:41.77262+00:00","source":"crossref"},{"observed_at":"2026-08-12T06:34:36.333875+00:00","source":"retraction_watch"}],"thesis":"As of 12 August 2026, this Paper Citation Record lists 100 of 119 outbound references and 0 inbound Pith citation observations for arXiv:2607.03784."}