{"as_of":"2026-08-08T10:10:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:959addad1fe651488c0635ecd5dca37664ec613c42bdaa8b66887ffcb50793e3","coverage":[{"denominator":46,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":46,"source":"paper_references, paper_reference_links","source_observed_at":"2026-07-31T14:14:14.863736Z","state":"measured"},{"denominator":46,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":46,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-08T06:32:00.761636+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/2607.24465/citation-record","integrity":"/paper/2607.24465/integrity","json":"/paper/2607.24465/citation-record.json","paper":"/paper/2607.24465"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-07-31T14:14:14.702179Z","title":"Food-101 – Mining Discriminative Components with Ran- dom Forests","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.702179Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:b7fc0107f2afef1af0682ef3292383876e5c424fe660a82190e89d9029b0e039","observation_id":"272283d6-3910-455b-a1f2-5dd58762c863","resolution":{"observed_at":"2026-07-31T14:14:14.702179Z","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":"2026-07-31T14:14:14.706754Z","title":"Remote sens- ing image scene classification: Benchmark and state of the art.Proceedings of the IEEE, 2017","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.706754Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:2e7564e4473c22f12f2acba5f93d7a9191ccfbef14450e5711995bb46a8a19a8","observation_id":"7d75456b-9837-48cd-a183-6987d79ec615","resolution":{"observed_at":"2026-07-31T14:14:14.706754Z","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":"2026-07-31T14:14:14.710539Z","title":"Describing textures in the wild","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.710539Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:a83839e0fec6dee58dd5e2ed0588436c7b39d7f59bd7632387a4c0e5772baf75","observation_id":"3ef0ae5c-e59b-452a-a30c-cb37af70953d","resolution":{"observed_at":"2026-07-31T14:14:14.710539Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1812.01718","last_updated":"2018-12-03T12:37:31Z","snapshot_observed_at":"2026-07-06T07:19:11.957225Z","submitted_at":"2018-12-03T12:37:31Z","title":"Deep Learning for Classical Japanese Literature","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1812.01718","snapshot_observed_at":"2026-07-31T14:14:14.714164Z","title":"Deep Learning for Classical Japanese Literature, 2018","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.714164Z"},"links":{"cited_paper":"/paper/1812.01718","citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:0910156da4895ac4fb38533e9c381b70795c1ffc8d153d8850f5f22cea9e8eb3","observation_id":"9d47f8e1-e740-44ef-a987-65cb7d272823","resolution":{"observed_at":"2026-07-31T14:14:14.714164Z","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":"2026-07-31T14:14:14.718675Z","title":"An Analysis of Single-Layer Networks in Unsupervised Feature Learn- ing","venue":null,"work_id":null,"year":2011},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.718675Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:035c9248e1d99b353e3c553f42b9729b023b06d61c11128d591be26ba33ba0ef","observation_id":"9857c403-88ff-4c6d-af1b-47a7ff79fc64","resolution":{"observed_at":"2026-07-31T14:14:14.718675Z","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":"2026-07-31T14:14:14.722252Z","title":"EMNIST: Extending MNIST to handwritten let- ters","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.722252Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:c9164adb8da01e1e23abb4b6175d23bb7654b278e718f92afe8bb8c31f0f5a7d","observation_id":"b206bf9b-d9f6-4d52-b8e0-9240ed06703b","resolution":{"observed_at":"2026-07-31T14:14:14.722252Z","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":"2026-07-31T14:14:14.726156Z","title":"Model breadcrumbs: Scaling multi-task model merging with sparse masks.Proceedings of the European Conference on Com- puter Vision, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.726156Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:1692513434357b8e77dca5143779e329066256a80456958886caf4d069ef7584","observation_id":"947df152-ba73-4e02-951a-ca166aff5e62","resolution":{"observed_at":"2026-07-31T14:14:14.726156Z","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":"2026-07-31T14:14:14.729488Z","title":"Roy, and Michael Carbin","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.729488Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:a498c41edbaaab2e15ab72404abb69bdfaf191cfb00e10f9c53d9c33dfb26186","observation_id":"c3493b9a-ad9d-4bb5-85c6-e2cc1060c9d8","resolution":{"observed_at":"2026-07-31T14:14:14.729488Z","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":"2026-07-31T14:14:14.732815Z","title":"Clip with generative latent replay: a strong baseline for in- cremental learning.BMVC, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.732815Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:254eed45681825255acb359fc152ec1030e91482b6684f5c1eb3a8c453131e2a","observation_id":"fa9348a3-932e-40d2-ba03-d16ae4ce4bf0","resolution":{"observed_at":"2026-07-31T14:14:14.732815Z","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":"2026-07-31T14:14:14.736080Z","title":"Task singular vectors: Reducing task in- terference in model merging.Proceedings of the IEEE con- ference on Computer Vision and Pattern Recognition, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.736080Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:80bdaebac01b37812fc0701f6558744fe2764e9ed2a59f5001e92bb399689298","observation_id":"4c7b3801-3579-434d-9f53-7ae1c202409b","resolution":{"observed_at":"2026-07-31T14:14:14.736080Z","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":"2026-07-31T14:14:14.739245Z","title":"Challenges in representation learning: A re- port on three machine learning contests.Neural Networks, 64, 2013","venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.739245Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:b8c8aeec356ad2d78ebf3209658950c2fc23c40f07f926f3898314228f5582ef","observation_id":"abc64c3e-4698-4586-b05a-7200e0a6d55f","resolution":{"observed_at":"2026-07-31T14:14:14.739245Z","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":"2026-07-31T14:14:14.742688Z","title":"EuroSAT: A Novel Dataset and Deep Learn- ing Benchmark for Land Use and Land Cover Classification","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.742688Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:6df0e79d45303729c79eeaf1df011a1bb7b20093fd3957743408715c084e3408","observation_id":"ebe00112-041b-4e53-a516-d00eb8bd8d5d","resolution":{"observed_at":"2026-07-31T14:14:14.742688Z","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":"2026-07-31T14:14:14.746107Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.746107Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:ba03a41f2a426c0e0b21f324b466c49fa19cc7ff38114c3f7991e356658d4dc0","observation_id":"85481ff5-4556-4e5e-8006-4e24e44d32eb","resolution":{"observed_at":"2026-07-31T14:14:14.746107Z","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":"2026-07-31T14:14:14.749468Z","title":"Editing models with task arithmetic","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.749468Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:9d49c7cc31c44b3a08ce05029cea7b4ab0737198bd82903af0fbdbe223eb0855","observation_id":"b55883c4-20b1-49b7-ab97-e86490d7cd5b","resolution":{"observed_at":"2026-07-31T14:14:14.749468Z","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":"2026-07-31T14:14:14.753660Z","title":"Vi- sual prompt tuning","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.753660Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:11d9401c86b6f6ae5dcd5543f71b059bca9074aa1bf633f07d868571052fc937","observation_id":"37a1c68d-1d33-45ac-a7e7-2a4827ba644a","resolution":{"observed_at":"2026-07-31T14:14:14.753660Z","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":"2026-07-31T14:14:14.757693Z","title":"Maple: Multi-modal prompt learning","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.757693Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:289d500d9458f4593d745b833e91dd24daf514296edace56bd66b1a791bd9a16","observation_id":"6543ad27-54bc-48a2-940b-738566ec5eca","resolution":{"observed_at":"2026-07-31T14:14:14.757693Z","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":"2026-07-31T14:14:14.761808Z","title":"Self-regulating prompts: Foundational model adaptation without forgetting","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.761808Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:6054154681cf5cdc898c899324b63a519b7b4d1ec7ee3f830b7117f48f6eb040","observation_id":"8055a3a7-9c2b-4ee8-a18d-cc682aaf9743","resolution":{"observed_at":"2026-07-31T14:14:14.761808Z","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":"2026-07-31T14:14:14.766087Z","title":"3d object representations for fine-grained categorization","venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.766087Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:54557ec599f76fa39e3109ee18ab5f7077d7041c7f4bf7f2eab77f63a9f0194b","observation_id":"2d4a4ce0-f836-4dab-8a88-6021f93c6037","resolution":{"observed_at":"2026-07-31T14:14:14.766087Z","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":"2026-07-31T14:14:14.769606Z","title":"Learning multiple layers of features from tiny images.University of Toronto,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.769606Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:c141d73b0e8438b5cc7e320bce25dbc1665052501854e8129af50e4e4fa508ed","observation_id":"7082b0ab-d5d0-412a-a12d-9633e63b8b7c","resolution":{"observed_at":"2026-07-31T14:14:14.769606Z","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":"2026-07-31T14:14:14.773247Z","title":"Mnist hand- written digit database.ATT Labs [Online]., 2, 2010","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.773247Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:fc4f48531dd8de3b32fb8531313b68ac6e0ebd1bae03789b30675d9738262a26","observation_id":"927b9ecc-5ecb-48f7-8189-a7578c6f0410","resolution":{"observed_at":"2026-07-31T14:14:14.773247Z","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":"2026-07-31T14:14:14.776783Z","title":"P-tuning: Prompt tuning can be comparable to fine-tuning across scales and tasks","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.776783Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:aef2c399cd17bd373d98f34facea7c89125c0982341b02e6cdbd0fe940c0b3fe","observation_id":"00cbfe90-c0f3-4345-98bf-e080dc7d4e1a","resolution":{"observed_at":"2026-07-31T14:14:14.776783Z","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":"2026-07-31T14:14:14.780295Z","title":"U-net transplant: the role of pre-training for model merging in 3d medical segmentation","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.780295Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:4b4ad97cb92f80e8cbe99e830cfcf1c0a605d1a09c8e8ee66cfb9ea5d469c933","observation_id":"f9d42dcf-4421-4bd7-9ecd-3ed3735d0fb2","resolution":{"observed_at":"2026-07-31T14:14:14.780295Z","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":"2026-07-31T14:14:14.783619Z","title":"MAGMAX: leveraging model merg- ing for seamless continual learning","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.783619Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:a616226a6caf2ff52e2fd75fa099f5fe844c256af712a63f1c42c818cc4bd6d2","observation_id":"1d468c15-8e56-4d28-a1db-b6d7e322e85d","resolution":{"observed_at":"2026-07-31T14:14:14.783619Z","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":"2026-07-31T14:14:14.786859Z","title":"Bagdanov, and Joost van de Weijer","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.786859Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:6e2a289a5f3b25a8c1dfa3cf070b971c40d2179a73dd89f62c2b5724bb5cc911","observation_id":"c5a6b957-cf4f-45b7-a344-1ba7b7de08dd","resolution":{"observed_at":"2026-07-31T14:14:14.786859Z","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":"2026-07-31T14:14:14.790122Z","title":"Merging models with fisher-weighted averaging","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.790122Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:219bb98ac3f691ade9c2c69dfe0ff903b6776f91991a748b7085587350f969f7","observation_id":"a8450735-c1fa-4d6a-9715-b328b3be9c65","resolution":{"observed_at":"2026-07-31T14:14:14.790122Z","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":"2026-07-31T14:14:14.793334Z","title":"An empirical investigation of the role of pre- training in lifelong learning.Journal of Machine Learning Research, 24(214):1–50, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.793334Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:bbd9f5411fc33bb35dc3d0a7b61ac3aeaea85b222f2f6a65519da95c8e224457","observation_id":"c4c59b97-55ae-4ef7-bc0d-d7284725df96","resolution":{"observed_at":"2026-07-31T14:14:14.793334Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2010.04495","last_updated":"2020-10-09T10:53:25Z","snapshot_observed_at":"2026-07-06T10:02:57.332438Z","submitted_at":"2020-10-09T10:53:25Z","title":"Linear Mode Connectivity in Multitask and Continual Learning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2010.04495","snapshot_observed_at":"2026-07-31T14:14:14.796450Z","title":"Linear mode connectivity in multitask and continual learning.arXiv preprint arXiv:2010.04495, 2020","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.796450Z"},"links":{"cited_paper":"/paper/2010.04495","citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:efd5925983a4870b29389f2cf9799a610cef73e169dffc8f8e33c8b60b43a81c","observation_id":"0c9c20c1-7e85-473b-8607-379ccf2f8d51","resolution":{"observed_at":"2026-07-31T14:14:14.796450Z","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":"2026-07-31T14:14:14.800769Z","title":"Reading digits in natural images with unsupervised feature learning","venue":null,"work_id":null,"year":2011},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.800769Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:d0a407d3a62a16005be3a87f83e61e517ff5c45a5f2de1b73082e582edbc7497","observation_id":"3f0ade5c-a901-4939-9ca8-242a8ac833a7","resolution":{"observed_at":"2026-07-31T14:14:14.800769Z","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":"2026-07-31T14:14:14.804519Z","title":"Automated Flower Classification over a Large Number of Classes","venue":null,"work_id":null,"year":2008},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.804519Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:df3604f2d72b58d86c8775171a82cb5f9d27e2d119cd556f5c8648de23257642","observation_id":"47fbe36b-418a-4603-bb43-3fe9fd16a4f1","resolution":{"observed_at":"2026-07-31T14:14:14.804519Z","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":"2026-07-31T14:14:14.807875Z","title":"Bag- danov, Simone Calderara, and Joost van de Weijer","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.807875Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:48b8a05507ec5b7275971384402c5747b5fb16931696364fe573baab1a10282d","observation_id":"a87aaddf-a58d-4c5d-861d-f6d27e4142f5","resolution":{"observed_at":"2026-07-31T14:14:14.807875Z","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":"2026-07-31T14:14:14.811530Z","title":"Parkhi, Andrea Vedaldi, Andrew Zisserman, and C","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.811530Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:8f88c99d5f95dc803857710b5d0dc0fb88486e7666dde1da3fe92f2b749d703a","observation_id":"94ba8af9-0517-4e7b-8a0e-39226817be8e","resolution":{"observed_at":"2026-07-31T14:14:14.811530Z","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":"2026-07-31T14:14:14.815594Z","title":"Learning transferable visual models from natural language supervision","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.815594Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:f2779463bb4993d1d833b98097e984d0a9e680a6d5f20780b132bcbe7085fadf","observation_id":"ed045f31-7c48-4b97-99e7-bbc3dd052994","resolution":{"observed_at":"2026-07-31T14:14:14.815594Z","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":"2026-07-31T14:14:14.819309Z","title":"Transporting task vectors across different architectures without training.International Conference on Machine Learning, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.819309Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:ee39761df744eb060e2c2c8b508f3825eaf6489239443f1f04568571e889a92f","observation_id":"a764d376-30b6-4cfa-b9ec-72cbc65043db","resolution":{"observed_at":"2026-07-31T14:14:14.819309Z","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":"2026-07-31T14:14:14.823112Z","title":"Manning, A","venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.823112Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:1a18772a29d8acbb52e856c43da5abca4c95754e376218cad8e32ab76b72fa8f","observation_id":"7f517fe6-b61c-46d4-a5b3-dee3d863397a","resolution":{"observed_at":"2026-07-31T14:14:14.823112Z","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":"2026-07-31T14:14:14.826361Z","title":null,"venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.826361Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:f9336712acec00396ed651b32cfe2d0387d282bf458a3a567575d1979cba2952","observation_id":"1fe1fb36-730d-4495-a1c5-f8e02dddd9eb","resolution":{"observed_at":"2026-07-31T14:14:14.826361Z","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":"2026-07-31T14:14:14.829720Z","title":"Model merging with svd to tie the knots.International Conference on Learning Repre- sentations, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.829720Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:174448006c03260dcd84974d270342c6fb3b3ca803ee1e75158879344e8cabd1","observation_id":"88bcfb14-fdc1-4a16-84c7-17db2d71b181","resolution":{"observed_at":"2026-07-31T14:14:14.829720Z","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":"2026-07-31T14:14:14.833296Z","title":"Veeling, Jasper Linmans, Jim Winkens, Taco Co- hen, and Max Welling","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.833296Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:b1a16edd6f685d616d3adaec0e332f5570fe384b03254239bf98068ed2dfde2b","observation_id":"ff0aad3e-7868-42f9-a4ae-7eca41ff5b69","resolution":{"observed_at":"2026-07-31T14:14:14.833296Z","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":"2026-07-31T14:14:14.836714Z","title":"Model soups: averaging weights of multi- ple fine-tuned models improves accuracy without increas- ing inference time","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.836714Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:3762f74b1d1ff2498e188e0979d78a16eba1a8ab3cd6a3ef0f54462e78eb23d4","observation_id":"42fdfbca-3a1b-46e3-86ae-700e8a82fda1","resolution":{"observed_at":"2026-07-31T14:14:14.836714Z","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":"2026-07-31T14:14:14.840291Z","title":"Fashion- mnist: a novel image dataset for benchmarking machine learning algorithms, 2017","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.840291Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:42dad935c51aef68665f1bcc1e4f711daa15d31e7b361290a68e63b9371ef6d9","observation_id":"98e0dd4f-b70b-4b6c-9ff3-108726c22407","resolution":{"observed_at":"2026-07-31T14:14:14.840291Z","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":"2026-07-31T14:14:14.843561Z","title":"Sun database: Exploring a large col- lection of scene categories.International Journal of Com- puter Vision, 2016","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.843561Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:89363fba8d923c1146a5c387264a7c9f92b98a0ee6bd5670a0947ddc8fdd089f","observation_id":"07bf9822-204a-4ec0-b138-4087915ae211","resolution":{"observed_at":"2026-07-31T14:14:14.843561Z","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":"2026-07-31T14:14:14.847130Z","title":"TIES-merging: Resolving interference when merging models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.847130Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:5837f13712d1a0cc54d63234579ebc446298ba2203157aeeb0f2e9b1cdaa83b1","observation_id":"be83c9e0-a05e-4540-8975-bb1f90b3e605","resolution":{"observed_at":"2026-07-31T14:14:14.847130Z","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":"2026-07-31T14:14:14.850354Z","title":"Adamerging: Adap- tive model merging for multi-task learning","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.850354Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:5c080a1ca8f38e1eef2a8db8edb02fed48f744369a1684f8a4c44e02c80138ed","observation_id":"ef37f863-20bb-4607-b320-04eeb0f87afb","resolution":{"observed_at":"2026-07-31T14:14:14.850354Z","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":"2026-07-31T14:14:14.853735Z","title":"Model merg- ing in llms, mllms, and beyond: Methods, theories, appli- cations, and opportunities.ACM Computing Surveys, 58(8): 1–41, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.853735Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:270c264ad1eaace2a89737463d6b975158fd2babf070434f1df67a0ce0208219","observation_id":"1317691c-7ad3-476a-83d4-b1592e92fd52","resolution":{"observed_at":"2026-07-31T14:14:14.853735Z","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":"2026-07-31T14:14:14.856885Z","title":"Conditional prompt learning for vision-language mod- els","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.856885Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:719712a41ea477edf355c3718be88cdfd4d76812c28be6c2b8cbae7184a3364e","observation_id":"68cbefc1-f95f-4c8e-a895-3200f706c377","resolution":{"observed_at":"2026-07-31T14:14:14.856885Z","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":"2026-07-31T14:14:14.860481Z","title":"Learning to prompt for vision-language models.Inter- national Journal of Computer Vision, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.860481Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:021162afe604ff21cc3dbff92d8ab16e43a3f8b3f4590677989bebd9cf5edbb7","observation_id":"a21bda87-142e-4f7a-8dcf-7c87833ea963","resolution":{"observed_at":"2026-07-31T14:14:14.860481Z","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":"2026-07-31T14:14:14.863736Z","title":"De- mystifying mergeability: Interpretable properties to predict model merging success.International Conference on Ma- chine Learning, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-07-31T14:14:14.863736Z"},"links":{"citing_paper":"/paper/2607.24465"},"observation_digest":"sha256:1caf3dc81e3a7bf8a469fceefdc86a70d25e02a65b2b50dd24ae64c938ef577a","observation_id":"11369e3e-a4ea-4eaf-92f2-a77d8b77bf53","resolution":{"observed_at":"2026-07-31T14:14:14.863736Z","resolver_source":null,"status":"malformed_identifier"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2607.24465","last_updated":"2026-07-27T14:04:48Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-03T05:36:59.540485Z","submitted_at":"2026-07-27T14:04:48Z","title":"Rethinking Expert Training for Model Merging with Prompt Learning"},"reference_resolution":{"displayed":46,"state_counts":{"malformed_identifier":1,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":45,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":46},"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":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"thesis":"As of 8 August 2026, this Paper Citation Record lists 46 of 46 outbound references and 0 inbound Pith citation observations for arXiv:2607.24465."}