{"as_of":"2026-08-10T21:18:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:d460376500b22dd03f747c836e9f134f359530277e390375ab5ef1b08ccac527","coverage":[{"denominator":43,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":43,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-10T14:26:11.903685Z","state":"measured"},{"denominator":47,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":47,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-10T06:31:04.303077+00:00","state":"measured"},{"denominator":4,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":4,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-01T19:55:52.417363Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-07-04T16:49:58.057648Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"cited_work":{"arxiv_id":"2501.15556","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2501.15556","snapshot_observed_at":"2026-07-04T16:49:58.057648Z","title":"arXiv preprint arXiv:2501.15556 (2025)","venue":null,"work_id":"a8c150a2-efaf-4826-a35a-9fcf6787d234","year":2025},"citing_paper":{"arxiv_id":"2507.15640","last_updated":"2026-04-12T09:28:58Z","snapshot_observed_at":"2026-08-09T03:59:56.664513Z","submitted_at":"2025-07-21T14:01:54Z","title":"Data Mixing Agent: Learning to Re-weight Domains for Continual Pre-training","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-05-19T03:36:50.366757Z"},"links":{"cited_paper":"/paper/2501.15556","citing_paper":"/paper/2507.15640"},"observation_digest":"sha256:4ebc6c46e7e196f43f854b4d78e52ed51bc749b1762ea38d262623f73bc66467","observation_id":"66f08c25-4762-406d-b8a7-017d9b59a568","resolution":{"observed_at":"2026-05-19T03:37:00.912798Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"cited_work":{"arxiv_id":"2501.15556","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2501.15556","snapshot_observed_at":"2026-07-04T16:49:58.057648Z","title":"arXiv preprint arXiv:2501.15556 (2025)","venue":null,"work_id":"a8c150a2-efaf-4826-a35a-9fcf6787d234","year":2025},"citing_paper":{"arxiv_id":"2606.24993","last_updated":"2026-06-23T15:39:38Z","snapshot_observed_at":"2026-08-08T04:49:38.395324Z","submitted_at":"2026-06-23T15:39:38Z","title":"The Geometry of Sequential Learning: Lie-Bracket Prediction of Transfer Order","version":1},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-06-26T00:11:51.793813Z"},"links":{"cited_paper":"/paper/2501.15556","citing_paper":"/paper/2606.24993"},"observation_digest":"sha256:626f6f54cbb2c80e2a4de19fde41f734733402519c33370026e0621c2db9e856","observation_id":"983c94c1-7b98-4d07-81c1-06320fa77607","resolution":{"observed_at":"2026-07-04T16:49:58.059064Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"cited_work":{"arxiv_id":"2501.15556","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2501.15556","snapshot_observed_at":"2026-07-04T16:49:58.057648Z","title":"arXiv preprint arXiv:2501.15556 (2025)","venue":null,"work_id":"a8c150a2-efaf-4826-a35a-9fcf6787d234","year":2025},"citing_paper":{"arxiv_id":"2606.29554","last_updated":"2026-06-28T18:44:08Z","snapshot_observed_at":"2026-07-07T00:03:31.715960Z","submitted_at":"2026-06-28T18:44:08Z","title":"Optimizer Memory Makes Shuffle Order a First-Order Source of Fine-Tuning Noise","version":1},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-06-30T07:15:56.974540Z"},"links":{"cited_paper":"/paper/2501.15556","citing_paper":"/paper/2606.29554"},"observation_digest":"sha256:718053142af60f025c655ce70bb8b2b9e0805cd3dee838b90803e859df7eb5ea","observation_id":"a4998d28-efb1-4b11-94a5-234f171e7fad","resolution":{"observed_at":"2026-06-30T07:24:21.838843Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.15556","snapshot_observed_at":"2026-08-01T19:55:52.417363Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.16821","last_updated":"2026-07-18T13:43:44Z","snapshot_observed_at":"2026-08-07T16:44:16.045474Z","submitted_at":"2026-07-18T13:43:44Z","title":"First-Order Predictable but Pairwise Fragile: Local Task Adaptation in Trained Transformers","version":1},"reference_index":2025,"source":"pdf_text","source_observed_at":"2026-08-01T19:55:52.417363Z"},"links":{"cited_paper":"/paper/2501.15556","citing_paper":"/paper/2607.16821"},"observation_digest":"sha256:bcf56e34bcf83671b88960afc172e9183aac683eaee8cef9f0692092ef8c839f","observation_id":"99e2bb68-a0c6-40a0-9d36-3d9d047fab04","resolution":{"observed_at":"2026-08-01T19:55:52.417363Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2501.15556/citation-record","integrity":"/paper/2501.15556/integrity","json":"/paper/2501.15556/citation-record.json","paper":"/paper/2501.15556"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"2309.16609","last_updated":"2023-09-28T17:07:49Z","snapshot_observed_at":"2026-08-09T21:25:20.369782Z","submitted_at":"2023-09-28T17:07:49Z","title":"Qwen Technical Report","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.16609","snapshot_observed_at":"2026-08-10T14:26:11.698242Z","title":"Qwen technical report","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.698242Z"},"links":{"cited_paper":"/paper/2309.16609","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:d5df5144fefe1e3a7e5f8883e31e3717932a90766bbf55e068201d22e55f38ac","observation_id":"486df368-2ddc-4efd-b476-b47eb9c754d6","resolution":{"observed_at":"2026-08-10T14:26:11.698242Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.558299Z","title":"An introduction to the geometry of stochastic flows","venue":null,"work_id":"3f9ccad4-b81c-42b1-bef3-c0d647138328","year":2004},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.704294Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:2f5e594b41a96700a9142b316ed6b3e39c6cf01ca7cd37047b37144be8a152cc","observation_id":"ed707237-6e6e-4032-aa6f-b8bbeb133d5e","resolution":{"observed_at":"2026-08-10T14:26:12.562997Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2207.11719","last_updated":"2023-07-09T21:53:45Z","snapshot_observed_at":"2026-08-07T21:40:40.556424Z","submitted_at":"2022-07-24T11:23:31Z","title":"Gradient-based Bi-level Optimization for Deep Learning: A Survey","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2207.11719","snapshot_observed_at":"2026-08-10T14:26:11.709561Z","title":"Gradient-based bi-level optimization for deep learning: A survey","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.709561Z"},"links":{"cited_paper":"/paper/2207.11719","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:3b63e70958782d3648e08df3ab65836f0e1b272a7a379cef31fbb9248ac24ec6","observation_id":"bac9eeaa-23ad-400b-8e3a-0ceee9d7fe4f","resolution":{"observed_at":"2026-08-10T14:26:11.709561Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.544454Z","title":"Gradnorm: Gra- dient normalization for adaptive loss balancing in deep multitask networks","venue":null,"work_id":"7519363c-e877-4960-ba85-b3c5e58b92d9","year":2018},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.715795Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:4fd08eeebd671a61edd94bceaf9306a6b49d0698df1c967449e52ae695f268f5","observation_id":"e37390be-1692-455f-a7a1-4498c8a91261","resolution":{"observed_at":"2026-08-10T14:26:12.549055Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.529017Z","title":"Just pick a sign: Optimizing deep multitask models with gradient sign dropout","venue":null,"work_id":"db8a3483-f5c9-427d-b3cb-18df842ba520","year":2020},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.725127Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:f268418442ff7a11055b86edc96218a8f517134ac06c4431e3ee9dedf1e88223","observation_id":"1f5a1101-34a6-47db-9988-354b8f9a166f","resolution":{"observed_at":"2026-08-10T14:26:12.534135Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.512722Z","title":"Order matters in the presence of dataset imbalance for multilingual learning","venue":null,"work_id":"88cbdeea-c556-4244-b454-cd5cb1dd9477","year":2024},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.730338Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:9c4ef626d040fa64b558dd95c0ebc0740ae3dd42c3956eedc840d64682d11eb8","observation_id":"f8871ea8-a030-4d93-88da-02395cec63b7","resolution":{"observed_at":"2026-08-10T14:26:12.517418Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1804.10200","last_updated":"2018-04-26T17:58:45Z","snapshot_observed_at":"2026-08-03T08:59:08.025421Z","submitted_at":"2018-04-26T17:58:45Z","title":"The loss landscape of overparameterized neural networks","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1804.10200","snapshot_observed_at":"2026-08-10T14:26:11.737232Z","title":"The loss landscape of overparameterized neural networks","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.737232Z"},"links":{"cited_paper":"/paper/1804.10200","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:89ae0beeca4055f6b5fa7992964fa450d0eca9f842b40c7512dff3566cadea1c","observation_id":"01c9ed4c-88f4-436b-96a8-94b21489a582","resolution":{"observed_at":"2026-08-10T14:26:11.737232Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.497479Z","title":"How to compute hessian- vector products? In The Third Blogpost Track at ICLR 2024, 2024","venue":null,"work_id":"58f5e8bd-8a6b-4d85-a94c-ef825f1b6b90","year":2024},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.742154Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:3dcd828e0c94ce9a8b658d05f29e56fb1eb04be27b5d35dc326f5a66e4026731","observation_id":"51ac65e2-4ee5-413e-ad42-cf9e2df54ea5","resolution":{"observed_at":"2026-08-10T14:26:12.502088Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.483416Z","title":"Laplace redux-effortless bayesian deep learning","venue":null,"work_id":"cb7e2049-7749-4568-966d-b9c755dfa88e","year":2021},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.746580Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:ba6b85343a680f4b7b0f3576d9f1056f80686a8d5e668f206fcae05be03715a0","observation_id":"2f1acc6f-42ca-4529-a8ff-9f9f040fe8da","resolution":{"observed_at":"2026-08-10T14:26:12.488155Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.468495Z","title":"Multiple-gradient descent algorithm (mgda) for multiobjective opti- mization","venue":null,"work_id":"d61f3760-9a58-459a-9bca-cad5fc989577","year":2012},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.750870Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:b0fa1add44515a3c0f3edae508b9e816f7de7fdb26bebd69ff2abb6ff20d5824","observation_id":"3b08a1e5-cce4-4471-8b65-ad8fb264fd30","resolution":{"observed_at":"2026-08-10T14:26:12.473162Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.454457Z","title":"Implicit biases in multitask and continual learningfrom a backward error analysis perspective","venue":null,"work_id":"b9a9c89c-394f-4481-9595-4bc6656ce7a4","year":2023},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.755115Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:c1f372feced95990529be843ea5bd0ec8df22c8c94ca50f2b9913950cd5c8d43","observation_id":"e1ef8831-db92-44bb-ae43-f5b850649534","resolution":{"observed_at":"2026-08-10T14:26:12.458983Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:11.760151Z","title":"Bilevel programming for hyperparameter optimization and meta-learning","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.760151Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:ce7db4812ced318f9e1fec08e549b1ac39ae54f225c63f4033300f31880aa0db","observation_id":"b3c73123-282e-4abd-9c0e-ba93ae54329c","resolution":{"observed_at":"2026-08-10T14:26:11.760151Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.430792Z","title":"Gradual domain adaptation: Theory and algorithms, 2023","venue":null,"work_id":"fbcee0b9-986e-429a-9259-3803e3d849e2","year":2023},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.764735Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:41c965fd8816d125b12240bc02ac0742f76ce339bb3a4300943cfbe271f96983","observation_id":"5e752474-8956-4b25-ab11-cce8fdcc7bde","resolution":{"observed_at":"2026-08-10T14:26:12.435452Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2103.02631","last_updated":"2022-02-16T11:20:05Z","snapshot_observed_at":"2026-08-10T14:56:29.520620Z","submitted_at":"2021-03-03T19:03:52Z","title":"RotoGrad: Gradient Homogenization in Multitask Learning","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2103.02631","snapshot_observed_at":"2026-08-10T14:26:11.769109Z","title":"Rotograd: Gradient homogenization in multitask learning","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.769109Z"},"links":{"cited_paper":"/paper/2103.02631","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:7ebfa755571d925dca32cf070bae5959ec4395ba3ceb8781b394af1748d965ac","observation_id":"23df4a3b-b424-4c68-b3ef-3cb0dc59ddbf","resolution":{"observed_at":"2026-08-10T14:26:11.769109Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.415323Z","title":"How does adaptive optimization impact local neural network geometry? Advances in Neural Information Processing Systems, 36, 2024","venue":null,"work_id":"dcbcc099-d4b0-46cf-a103-6ecb8d177ad7","year":2024},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.774403Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:154d077aa2b8d017c5011b78cd2a0927e640f9f547aae2c7a7982af67bd973d7","observation_id":"f1af2ce6-ae7c-47ec-973c-862e3f724028","resolution":{"observed_at":"2026-08-10T14:26:12.420858Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1412.6980","last_updated":"2017-01-30T01:27:54Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2014-12-22T13:54:29Z","title":"Adam: A Method for Stochastic Optimization","version":9},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1412.6980","snapshot_observed_at":"2026-08-10T14:26:11.778674Z","title":"Adam: A method for stochastic optimization","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.778674Z"},"links":{"cited_paper":"/paper/1412.6980","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:3978fc70c0cb49c278c8046c404854371033aa1e183f5104affa39755514db37","observation_id":"6f19d429-6a4c-4978-8f6b-3ef1df4153fb","resolution":{"observed_at":"2026-08-10T14:26:11.778674Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:11.783112Z","title":"Understanding black-box predictions via influence functions","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.783112Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:955d2a72171792741c105e213a889fd4fc2fe1b0a53b05df6194e4fa645ba3d1","observation_id":"d0a9e500-c4dc-4d8f-859e-93b7a4df8bf2","resolution":{"observed_at":"2026-08-10T14:26:11.783112Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.387756Z","title":"Ubernet: Training a universal convolutional neural network for low-, mid-, and high-level vision using diverse datasets and limited memory","venue":null,"work_id":"ce3928d0-0d42-4ce3-9365-c471471d4adb","year":2017},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.787610Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:e545582c01a0b6e39584338ee9ac2031a2072252d49b879b0f9152814086ba31","observation_id":"daafc727-b59c-4625-ae4b-20c35e363224","resolution":{"observed_at":"2026-08-10T14:26:12.393077Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2002.11361","last_updated":"2020-02-26T08:59:40Z","snapshot_observed_at":"2026-07-06T09:00:12.600151Z","submitted_at":"2020-02-26T08:59:40Z","title":"Understanding Self-Training for Gradual Domain Adaptation","version":1},"cited_work":{"arxiv_id":"2002.11361","doi":null,"metadata_source":"pith","pith_arxiv_id":"2002.11361","snapshot_observed_at":"2026-08-10T14:26:12.004949Z","title":"Understanding Self-Training for Gradual Domain Adaptation","venue":"cs.LG","work_id":"06cff8c4-8d35-4487-b51c-1b3347ba1391","year":2020},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.792189Z"},"links":{"cited_paper":"/paper/2002.11361","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:d982a51a0845a17f249b46f7ab0b10217751a668352974c5620dd5357262eb1a","observation_id":"65624969-7e5a-4f44-acb0-189db6b01805","resolution":{"observed_at":"2026-08-10T14:26:12.011585Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:11.796835Z","title":"Smooth manifolds","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.796835Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:20a9b7636e0f4039009be31aa1d493d27203013247f7ed44b3870b815594228e","observation_id":"324df71c-ff94-45cc-b056-0832ea3a4256","resolution":{"observed_at":"2026-08-10T14:26:11.796835Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.359961Z","title":"Sequential reptile: Inter-task gradient alignment for multilingual learning","venue":null,"work_id":"141b597e-36a8-4b07-b70f-1b17b7cd42fd","year":2022},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.801067Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:448a051534a1f43b503b61239b88d79cdc648fe3f58a350d6477c79a8a258d4f","observation_id":"45bd61c8-5f83-41ae-8939-13f71a54a450","resolution":{"observed_at":"2026-08-10T14:26:12.364697Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.345404Z","title":"Pareto multi-task learning","venue":null,"work_id":"04b638bd-cda5-4e3b-bc01-e24feec4a7a2","year":2019},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.805332Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:d3db153f0213fb6b5527d98b3d93cf18effb2b2179663658a29d8568c0d9d5fa","observation_id":"eadc5bfc-d4e0-4a1e-80d2-ade2bf1353fd","resolution":{"observed_at":"2026-08-10T14:26:12.349971Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:11.809976Z","title":"On the limited memory bfgs method for large scale optimization","venue":null,"work_id":null,"year":1989},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.809976Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:eeb6854bd76e26d01b9e4132e5e9d9e8b514b2ea6c6055f3c4da0476b7c0b43b","observation_id":"876267d2-d9b9-4dd2-867f-8d2b5e7f7cef","resolution":{"observed_at":"2026-08-10T14:26:11.809976Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.322170Z","title":"Towards impartial multi-task learning","venue":null,"work_id":"f58102e1-9271-43fe-bab2-1f7b8589f53f","year":2021},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.814511Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:2ac4295633f65d827c9a9258544f4b693c9d3e1264861826ff2ff19337cd5758","observation_id":"4f6144cb-c559-43ab-9d3b-70f0b33fd8ed","resolution":{"observed_at":"2026-08-10T14:26:12.326801Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1711.05101","last_updated":"2019-01-04T21:01:49Z","snapshot_observed_at":"2026-08-09T20:34:52.923500Z","submitted_at":"2017-11-14T14:24:06Z","title":"Decoupled Weight Decay Regularization","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1711.05101","snapshot_observed_at":"2026-08-10T14:26:11.819289Z","title":"Decoupled weight decay regularization","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.819289Z"},"links":{"cited_paper":"/paper/1711.05101","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:b6406a1b704e37eced0d4ad4811ac11d2d7fe7c39982e77a128e77f5c723fbf4","observation_id":"974885eb-75c3-45fb-97d6-d4613332fa3c","resolution":{"observed_at":"2026-08-10T14:26:11.819289Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.307743Z","title":"Abide by the law and follow the flow: Conservation laws for gradient flows","venue":null,"work_id":"1cbb72d4-57a4-4c77-90f6-4e02b5ba6461","year":2024},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.825356Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:798da8730b8120c420528f8b007f77bab5e8e53e3465b18924c035fcf6838a1b","observation_id":"18526ae6-2418-40dd-8a06-7ac765ce7b29","resolution":{"observed_at":"2026-08-10T14:26:12.312456Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.293415Z","title":"The benefit of multitask representation learning","venue":null,"work_id":"bca51b88-cf8e-4d9d-a79d-2907d74ae726","year":2016},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.830162Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:732c361ad7b5602cbbaa88bf96129118c1389acd5fc224824993c0d8f7b7c008","observation_id":"9a539a25-55ae-40a1-a2c4-bd78b3d1735b","resolution":{"observed_at":"2026-08-10T14:26:12.298587Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.279105Z","title":"Multi-task learning as a bargaining game","venue":null,"work_id":"06ea0420-d67b-4c8c-83ae-5427811a3c00","year":2022},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.835174Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:d5658f28ca7786fe308cb3c089ce54ff1128d270d2713d35cd72d94c25261e35","observation_id":"0c77dd37-8202-4087-9460-b43c84af7637","resolution":{"observed_at":"2026-08-10T14:26:12.283668Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:11.840452Z","title":"Numerical optimization","venue":null,"work_id":null,"year":1999},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.840452Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:569aa2b36ce90242bad41712a1a83d8a68f6d001266be74e434418695bde5b14","observation_id":"0aaf6509-978b-49cc-8bd8-7d7d7f4e71b8","resolution":{"observed_at":"2026-08-10T14:26:11.840452Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.254730Z","title":"Fast exact multiplication by the hessian","venue":null,"work_id":"ce6e268b-9cb9-4bfb-8776-54a954dcb2fc","year":1994},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.845036Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:bdf24004120f43b557849c7d7c4e1bff6b94a6b65c47531c60dcb82875bf12f7","observation_id":"d4c1ee67-aa4b-409d-ba47-df899be961f5","resolution":{"observed_at":"2026-08-10T14:26:12.259569Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:11.849578Z","title":"Estimating training data influence by tracing gradient descent","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.849578Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:58a4d365f93442611402afd34722b0647370c475a0a7868cd12f221e304cc195","observation_id":"8ab81ee9-19e2-40d3-b02f-e37f9ab10729","resolution":{"observed_at":"2026-08-10T14:26:11.849578Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:11.854466Z","title":"Exploring the limits of transfer learning with a unified text-to-text transformer","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.854466Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:4e53a23a21de25316ba05dca57d3916a7d3465d937b4aed8522b1fff25b35084","observation_id":"a89ea3d2-aed5-45e1-8de5-cdde2a5f1474","resolution":{"observed_at":"2026-08-10T14:26:11.854466Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.219901Z","title":"Scalarization for multi-task and multi-domain learning at scale","venue":null,"work_id":"3b6da57e-e074-428f-8322-1fe830e7a5e2","year":2024},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.859010Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:c64ea020327c865337a88d226e25b48f71d0b4332ba5a42f6d17b14ddb56677e","observation_id":"07405d1c-48f6-47c9-a460-65f46f11835d","resolution":{"observed_at":"2026-08-10T14:26:12.224655Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.204027Z","title":"Scaling up influence functions","venue":null,"work_id":"26a79bf2-9b43-48e1-a7b8-a9ec3ccc8deb","year":2022},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.863417Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:ade1c9bc20a8b9f2b0b37b4efc4bf51ee39de1e58e1518fa98385ba3c72c0180","observation_id":"8b1d10c1-4199-4550-8364-4e7bc9377438","resolution":{"observed_at":"2026-08-10T14:26:12.209513Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2307.09288","last_updated":"2023-07-19T17:08:59Z","snapshot_observed_at":"2026-08-07T12:56:43.323460Z","submitted_at":"2023-07-18T14:31:57Z","title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2307.09288","snapshot_observed_at":"2026-08-10T14:26:11.868025Z","title":"Llama 2: Open foundation and fine-tuned chat models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.868025Z"},"links":{"cited_paper":"/paper/2307.09288","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:449758f8c4342a8eee060d06669d6fca276ae5d00d57068e9b8d59e922427963","observation_id":"133cc434-fc3b-4a83-952b-3bfec5a37bbe","resolution":{"observed_at":"2026-08-10T14:26:11.868025Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.185526Z","title":"Gradient vaccine: Investigating and improving multi-task optimization in massively multilingual models","venue":null,"work_id":"87310cbe-a65e-469c-9890-4a2a25aac06c","year":2021},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.873054Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:84ba4473452d59226021e89d9572f15bd055da9c012b2f662a759294fc1a9e39","observation_id":"8aa28f08-fc33-451f-8fb1-05e6c8a83787","resolution":{"observed_at":"2026-08-10T14:26:12.192435Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2201.13011","last_updated":"2022-08-01T12:08:24Z","snapshot_observed_at":"2026-08-05T10:35:32.786929Z","submitted_at":"2022-01-31T06:04:47Z","title":"On the Power-Law Hessian Spectrums in Deep Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2201.13011","snapshot_observed_at":"2026-08-10T14:26:11.877550Z","title":"On the power-law hessian spectrums in deep learning","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.877550Z"},"links":{"cited_paper":"/paper/2201.13011","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:71ed1e3ec474eac044780faaae9c3036efe76068c3cf2e0ec43a2ed8ae6cf4cd","observation_id":"75c91cea-3438-4f2d-822e-be51d2d5c310","resolution":{"observed_at":"2026-08-10T14:26:11.877550Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2010.11934","last_updated":"2021-03-11T18:45:13Z","snapshot_observed_at":"2026-08-09T05:32:57.900052Z","submitted_at":"2020-10-22T17:58:14Z","title":"mT5: A massively multilingual pre-trained text-to-text transformer","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2010.11934","snapshot_observed_at":"2026-08-10T14:26:11.882025Z","title":"mt5: A massively multilingual pre-trained text-to-text transformer","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.882025Z"},"links":{"cited_paper":"/paper/2010.11934","citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:957c41fdf2c8fba78a39b337e63ff094693a4992792c447692e8e8dabf31fe32","observation_id":"09b94bc8-dde0-43fd-b8fc-76d482a9eb51","resolution":{"observed_at":"2026-08-10T14:26:11.882025Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.169662Z","title":"Adatask: A task-aware adaptive learning rate approach to multi-task learning","venue":null,"work_id":"6ea2677d-a71b-4330-a53d-212e8863ae89","year":2023},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.886429Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:cff737e8bc5d9753a7877268e54b23135e065a97af0e1ae7d9797578ed2cac83","observation_id":"e5f937f5-4782-4b03-81f5-3590b0083346","resolution":{"observed_at":"2026-08-10T14:26:12.174319Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.155421Z","title":"Gradient surgery for multi-task learning","venue":null,"work_id":"190d786a-f9cf-423f-946e-d0dac04ad0a3","year":2020},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.890582Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:b5945412459f14e3a98a0bae89175107ec4c61135f588e586967fe23bdb182a4","observation_id":"43bd8547-1d03-4069-8a85-cb8d5053e21d","resolution":{"observed_at":"2026-08-10T14:26:12.160100Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.140458Z","title":"Gradual domain adaptation via gradient flow","venue":null,"work_id":"f498ae88-cd4f-41f1-bca8-b666c1ad8cc7","year":2024},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.894721Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:ac7581b139c08cbc911c5144a0f1edfff51e386d4e13049a486488f943aca376","observation_id":"b0ae2db7-337f-4773-b95a-9063ee078f7d","resolution":{"observed_at":"2026-08-10T14:26:12.145462Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.125020Z","title":null,"venue":null,"work_id":"f6b30c24-fe2a-4408-b6c4-0ff1de2db1cd","year":null},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.899008Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:d582ea8bc0bd0f0a5ad38900011fdbaf68bd12f1d587688f05be1940ca8e40c9","observation_id":"fbbad324-0496-4302-b887-68fac5fa22b8","resolution":{"observed_at":"2026-08-10T14:26:12.129720Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:12.106375Z","title":null,"venue":null,"work_id":"8f117c36-ad9d-4b3b-a8d9-9219e6918c96","year":2000},"citing_paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:11.903685Z"},"links":{"citing_paper":"/paper/2501.15556"},"observation_digest":"sha256:049d253a775ddb6dfe2fd914ee8a74de47b8b0204d1db7a65ed6db32d4abb6a3","observation_id":"a0bc995a-daaf-496d-a18e-c857bbc54144","resolution":{"observed_at":"2026-08-10T14:26:12.112756Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2501.15556","last_updated":"2025-01-26T15:12:06Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-10T14:08:22.206644Z","submitted_at":"2025-01-26T15:12:06Z","title":"Commute Your Domains: Trajectory Optimality Criterion for Multi-Domain Learning"},"reference_resolution":{"displayed":43,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":18,"verified_exact":1,"verified_fuzzy":24},"total_outbound_references":43},"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":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"thesis":"As of 10 August 2026, this Paper Citation Record lists 43 of 43 outbound references and 4 inbound Pith citation observations for arXiv:2501.15556."}