{"as_of":"2026-08-10T20:39:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:190b17e3e6d7430210be9f90f2b0bd887e29b74e46c145b0cc11e99ee361f171","coverage":[{"denominator":57,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":57,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-03T10:52:03.139813Z","state":"measured"},{"denominator":57,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":57,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-10T06:31:04.303077+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/2601.08617/citation-record","integrity":"/paper/2601.08617/integrity","json":"/paper/2601.08617/citation-record.json","paper":"/paper/2601.08617"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-03T10:52:02.781278Z","title":"Flamingo: a visual language model for few-shot learning","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.781278Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:2c801562e538b6591d03b237ed150e52d8fc57fb87d7bca5cbddd8cc2683d0e6","observation_id":"d361444d-c4a6-4b78-8210-e9584b9e55aa","resolution":{"observed_at":"2026-08-03T10:52:02.781278Z","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":"2026-08-03T10:52:02.793669Z","title":"Food-101 – mining discriminative components with random forests","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.793669Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:df6c951c56c012392eae7d642c10486f7835f5b4fc5217c324b0ec164956a1b6","observation_id":"1ea7cd30-2200-4479-bc50-315288888433","resolution":{"observed_at":"2026-08-03T10:52:02.793669Z","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":"2026-08-03T10:52:02.799887Z","title":"Calibrating deep neural networks by pairwise constraints","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.799887Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:c497640bb22dae04c19b1b70240303b1eec1f55b806aab65efb8b347480a5b53","observation_id":"045e2953-f384-442e-8f76-413e724b830b","resolution":{"observed_at":"2026-08-03T10:52:02.799887Z","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":"2026-08-03T10:52:02.805314Z","title":"Cimpoi, S","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.805314Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:922d83cf44058c9ec9a48f395ad2b350dfe0e51ab59a3e99898f1c8e070aa79d","observation_id":"d2a6673d-0020-4e4a-92e7-c48628e5cf43","resolution":{"observed_at":"2026-08-03T10:52:02.805314Z","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":"2026-08-03T10:52:02.811459Z","title":"Imagenet: A large-scale hierarchical image database","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.811459Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:4ac9c140320b257c3198dcb3a9e2fa05989388a4b75a757584961c27d40a1c1c","observation_id":"4721f8a8-30e6-445f-b587-2a4763571608","resolution":{"observed_at":"2026-08-03T10:52:02.811459Z","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":"2026-08-03T10:52:02.816261Z","title":"Weinberger","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.816261Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:dbb6e19447ee645694bc4448b5b4c8ba902ccbb113692cf4e35bb89e1a48e696","observation_id":"cf4a0450-f6b8-4290-95e9-81445028f932","resolution":{"observed_at":"2026-08-03T10:52:02.816261Z","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":"2026-08-03T10:52:02.821553Z","title":"Visual-language prompt tuning with knowledge-guided context optimization","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.821553Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:3c15a9e863c340dd0038320c365fbe9b6e10743cb51fcd204d659d2551bd47aa","observation_id":"79939933-fa7f-4692-b8ee-ea9ca6d69b91","resolution":{"observed_at":"2026-08-03T10:52:02.821553Z","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":"2026-08-03T10:52:02.826662Z","title":null,"venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.826662Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:2138489c7aa4232154c6cd8cb863c6fa723fe842d2d7b38fcd92e39b2821a53f","observation_id":"81b4b4d1-ece4-44fb-b7be-75f9cd3f55e6","resolution":{"observed_at":"2026-08-03T10:52:02.826662Z","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":"2026-08-03T10:52:02.834031Z","title":"Cubuk, Barret Zoph, Justin Gilmer, and Balaji Lakshminarayanan","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.834031Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:8b2261c1bb311d58d4d8aeb10969da6d33f330eb0125e70b2f981d170c5ef635","observation_id":"da7fcd24-984c-4ec9-b6ed-53f332f2a695","resolution":{"observed_at":"2026-08-03T10:52:02.834031Z","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":"2026-08-03T10:52:02.839125Z","title":"The many faces of robustness: A critical analysis of out-of-distribution generalization.ICCV, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.839125Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:9c424adfaf96bd05e3ad81a3a422ad9cc52cd6e5b803e8b7356ba2d7b645f9b2","observation_id":"83504676-afc5-4407-8753-dcc5a956988b","resolution":{"observed_at":"2026-08-03T10:52:02.839125Z","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":"2026-08-03T10:52:02.845688Z","title":"Natural adversarial examples","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.845688Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:9574496378876cabaf13a211d7077206f539c78aec94c1a294a97e0a5fd20c7f","observation_id":"2b6739d0-5b16-4c71-840c-f947a9dca61b","resolution":{"observed_at":"2026-08-03T10:52:02.845688Z","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":"2026-08-03T10:52:02.856613Z","title":null,"venue":null,"work_id":null,"year":1964},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.856613Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:026b0469d0ca07c48bdfe4b58d6729f09702f177f850a3855df94ca3f9de6b19","observation_id":"e6c05e22-2179-4397-8814-88459749bbe9","resolution":{"observed_at":"2026-08-03T10:52:02.856613Z","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":"2026-08-03T10:52:02.864238Z","title":"Le, Yun-Hsuan Sung, Zhen Li, and Tom Duerig","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.864238Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:c11f675b657b9dbac511fb0958bd33259b049a05e2db8294406511ec5b24061a","observation_id":"25a7da41-d0cf-417c-9254-49dd1d1bdd94","resolution":{"observed_at":"2026-08-03T10:52:02.864238Z","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":"2026-08-03T10:52:02.869562Z","title":"Simple but effective: CLIP embed- dings for embodied AI","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.869562Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:04439d954e2275c3eeff824d2b1e54c1a52f475a20ff1d52610cd030854d9e83","observation_id":"e2b17526-01af-4a60-9af4-70678241b66c","resolution":{"observed_at":"2026-08-03T10:52:02.869562Z","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":"2026-08-03T10:52:02.878387Z","title":"Biomedcoop: Learning to prompt for biomedi- cal vision-language models","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.878387Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:80bab69ac2df175c84a555ecc86ebdc3a597dc86bc867ab4caec66cac2b1d3e4","observation_id":"044e3057-d060-4132-a237-c5ef0db01c93","resolution":{"observed_at":"2026-08-03T10:52:02.878387Z","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":"2026-08-03T10:52:02.884957Z","title":"3d object representations for fine-grained categorization","venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.884957Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:0c56a7d9856f7505b3a9a1ec6be7480c3466e6ff904dacfbd7baa8f2dcf5ae90","observation_id":"8e042156-b85c-444b-a865-b88292ebf123","resolution":{"observed_at":"2026-08-03T10:52:02.884957Z","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":"2026-08-03T10:52:02.890945Z","title":"One-shot learning of object categories.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2006","venue":null,"work_id":null,"year":2006},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.890945Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:e6e26cce8650e0ba1c6bea29522f929c2aa0330a57cfecc3256a8a756151140f","observation_id":"312ddf1f-d406-43ad-ace1-cdb085888f77","resolution":{"observed_at":"2026-08-03T10:52:02.890945Z","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":"2026-08-03T10:52:02.897211Z","title":"Selvaraju, Akhilesh Deepak Got- mare, Shafiq Joty, Caiming Xiong, and Steven Hoi","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.897211Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:9e8870e1d00324e94bf623f59337a804ffa596bf06d74af7ca709eff3c3f1afd","observation_id":"9e472948-1b49-43b9-a028-f32c93d05c8d","resolution":{"observed_at":"2026-08-03T10:52:02.897211Z","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":"2026-08-03T10:52:02.905086Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.905086Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:2a9916c5947800d0b579221352c69199996cddce5553dfe4b37433c2074c0b3b","observation_id":"5abd3aa1-c9cf-4d4a-a9aa-8f0ef7c6a492","resolution":{"observed_at":"2026-08-03T10:52:02.905086Z","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":"2026-08-03T10:52:02.913299Z","title":"CLIP-reID: exploiting vision-language model for image re-identification without concrete text labels","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.913299Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:5414cb801291d084e2d3dbf2a3739f98fbc27a5ce7c34e83d53523c7483110a9","observation_id":"a0a65c92-5f91-416f-8dc5-f7861a8f4f67","resolution":{"observed_at":"2026-08-03T10:52:02.913299Z","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":"2026-08-03T10:52:02.920019Z","title":"The devil is in the margin: Margin-based label smooth- ing for network calibration","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.920019Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:9cb39ace346ad722a8d63f9cba5ad144d06f1edb58809ef331b6ced624ffb831","observation_id":"3b7fcbf6-6cf5-4159-b862-18deef90a85c","resolution":{"observed_at":"2026-08-03T10:52:02.920019Z","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":"2026-08-03T10:52:02.927114Z","title":"Class adaptive network calibration","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.927114Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:d295ad62205b2590cf8b9a36afcaf8d197f95a3d870eb1ff2d06cf34cfe2ad36","observation_id":"95f6eb96-c75d-4532-91f4-2db16adfa440","resolution":{"observed_at":"2026-08-03T10:52:02.927114Z","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":"2026-08-03T10:52:02.933080Z","title":"Visual instruction tuning, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.933080Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:2cec0dc3175c67eda70118898db177c4bc0d64ea55ce61a943576b29381d7286","observation_id":"2d38fb82-fe67-4399-9508-44d0234d13c3","resolution":{"observed_at":"2026-08-03T10:52:02.933080Z","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":"2026-08-03T10:52:02.941206Z","title":"Decoupled weight de- cay regularization","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.941206Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:9d3c903c1a7b05bc069d0ad5beefae2e9a5d068286f0e0ca1e15f5887b13b23c","observation_id":"c1551147-339c-44e1-b2c2-c6ae23669b2b","resolution":{"observed_at":"2026-08-03T10:52:02.941206Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.19060","last_updated":"2025-02-05T00:47:41Z","snapshot_observed_at":"2026-08-10T12:46:02.070673Z","submitted_at":"2025-01-31T11:47:15Z","title":"Contrast-Aware Calibration for Fine-Tuned CLIP: Leveraging Image-Text Alignment","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.19060","snapshot_observed_at":"2026-08-03T10:52:02.947578Z","title":"Contrast-aware calibration for fine-tuned CLIP: Leveraging image-text alignment.arXiv preprint arXiv:2501.19060, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.947578Z"},"links":{"cited_paper":"/paper/2501.19060","citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:8d829dc85124f38879a4e428b3b7f82cbffa1d2529e15fe1ea147f56091c655a","observation_id":"7983b418-208c-47f3-b0ef-7e8c572b5c25","resolution":{"observed_at":"2026-08-03T10:52:02.947578Z","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":"2026-08-03T10:52:02.954536Z","title":null,"venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.954536Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:8f9b5bd34cc0f1d82365f8c3cd0e1e753108be508a4f66152ecd190509967eec","observation_id":"4416cb08-c621-438f-b758-af53985ec52c","resolution":{"observed_at":"2026-08-03T10:52:02.954536Z","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":"2026-08-03T10:52:02.962052Z","title":"Test- time prompt tuning for zero-shot generalization in vision- language models","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.962052Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:ab14d7938114cffb53c0f9b2fbf59b11f6ef0bf808b840152b6587c5016499aa","observation_id":"5561ff0d-4d35-4f84-9250-472eec0c7c0c","resolution":{"observed_at":"2026-08-03T10:52:02.962052Z","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":"2026-08-03T10:52:02.967518Z","title":"Revisiting the calibration of modern neu- ral networks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.967518Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:2b42fb9f84428f3b9b3d0d0de47c6da3a3f8606109c09868452a696b7bfe5690","observation_id":"1cafa23b-bfc5-4533-809d-37dba075f64a","resolution":{"observed_at":"2026-08-03T10:52:02.967518Z","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":"2026-08-03T10:52:02.973447Z","title":"Bayesadapter: enhanced uncertainty estimation in CLIP few-shot adaptation.International Journal of Computer Vision (IJCV), 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.973447Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:aa8f816e498ff6156cdd1abd342130eb0531a26b1a0c99400f132a2ae36c32d6","observation_id":"aa02ad61-e1ef-4b83-a859-4f4c3e6f5498","resolution":{"observed_at":"2026-08-03T10:52:02.973447Z","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":"2026-08-03T10:52:02.983946Z","title":"Slip: Self-supervision meets language-image pre- training","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.983946Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:9a33585d3346e96b1cf1394435071c445bc7a0b8848a9e2a94ca6a271a43c216","observation_id":"c9354d2b-d7c0-4236-81ab-792c386dd73f","resolution":{"observed_at":"2026-08-03T10:52:02.983946Z","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":"2026-08-03T10:52:02.990101Z","title":"Calibrating deep neural networks using focal loss","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.990101Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:4699518dca5f86de231b307304ff833109505604d3465bec5e53432464c469bc","observation_id":"45d8ccf9-5aa5-4211-80f8-1322a66f47d7","resolution":{"observed_at":"2026-08-03T10:52:02.990101Z","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":"2026-08-03T10:52:02.995501Z","title":"When does label smoothing help? NeurIPS, 32, 2019","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.995501Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:c24a6d4dcc10a5f9b1fbf50eaab023919d95a3f805b7f145a252b6e4e8489cd4","observation_id":"7d8c9d16-375b-4e83-8719-f3afdadf9b86","resolution":{"observed_at":"2026-08-03T10:52:02.995501Z","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":"2026-08-03T10:52:03.000137Z","title":"Robust calibration of large vision- language adapters","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.000137Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:c5ea4b2e2b3e09b55719b48fca74432ae27ab8de9e4b1b1db847e39ec60b79cc","observation_id":"0599720d-135d-472e-aa1e-dfdbf34e1bcf","resolution":{"observed_at":"2026-08-03T10:52:03.000137Z","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":"2026-08-03T10:52:03.004921Z","title":"Nilsback and A","venue":null,"work_id":null,"year":2008},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.004921Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:485bf530ef5d441991cc18e47bd7826db188cfc7c58a72f4c02ecb6f194d72fc","observation_id":"e18fbb7d-341d-4cc9-b60a-fc345c2ec851","resolution":{"observed_at":"2026-08-03T10:52:03.004921Z","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":"2026-08-03T10:52:03.009564Z","title":"Towards calibrated robust fine- tuning of vision-language models.Advances in Neural Infor- mation Processing Systems, 37:12677–12707, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.009564Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:079ee76c8880d62e3f05bf36cd37c1bb88eb4d10576ac970804aa68fcd3eede6","observation_id":"815db61f-4446-4c6e-a3db-09e45651308a","resolution":{"observed_at":"2026-08-03T10:52:03.009564Z","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":"2026-08-03T10:52:03.014028Z","title":"VLP: Vision language planning for autonomous driv- ing","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.014028Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:ac32563568a36dbb12e9eb580c404e6b9ed852411bda3e4046e8bae693535c26","observation_id":"1e6193cd-981e-45e3-9b7e-5bc42a1dd700","resolution":{"observed_at":"2026-08-03T10:52:03.014028Z","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":"2026-08-03T10:52:03.018700Z","title":null,"venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.018700Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:cde09975ccaecba74e9bf1dbfbca0f96736e2dda13d96904bb6919e83cd3af6b","observation_id":"39988803-00eb-45d1-b573-3ab1c713f483","resolution":{"observed_at":"2026-08-03T10:52:03.018700Z","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":"2026-08-03T10:52:03.023393Z","title":"Regularizing neural networks by pe- nalizing confident output distributions","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.023393Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:72ed0ddba6b50866e6fdfc03b4a8e0b835eb02a62901e941d878f32c4172f545","observation_id":"cc4367c8-26d9-48d1-8784-d32a0b82266d","resolution":{"observed_at":"2026-08-03T10:52:03.023393Z","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":"2026-08-03T10:52:03.028763Z","title":"Learning transferable visual models from natural language supervision","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.028763Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:c61deeb5e9970fd32e1b1afa8a75595fd614b71168a8c90313128194d63761b7","observation_id":"0fc01866-c275-4033-bf14-cb9afdaa8395","resolution":{"observed_at":"2026-08-03T10:52:03.028763Z","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":"2026-08-03T10:52:03.035386Z","title":"Do ImageNet classifiers generalize to Ima- geNet? InProceedings of the 36th International Conference on Machine Learning, pages 5389–5400","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.035386Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:306283fb036a3ca5d01d3cc4a19f038ed1a4e79bdeebf52e8377fe8914dac937","observation_id":"1fdb4f94-26b2-4767-b2dc-5f370d00ea0a","resolution":{"observed_at":"2026-08-03T10:52:03.035386Z","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":"2026-08-03T10:52:03.041165Z","title":"O-TPT: Orthogonality constraints for calibrating test-time prompt tuning in vision-language models","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.041165Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:0dda6bc3665a4e25eb546fe00ff80cda151ba58ac57413a839537f857fe4c0ab","observation_id":"7affa219-5227-4e61-96da-e1be17a35ca9","resolution":{"observed_at":"2026-08-03T10:52:03.041165Z","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":"2026-08-03T10:52:03.047061Z","title":"Test- time prompt tuning for zero-shot generalization in vision- language models.Advances in Neural Information Process- ing Systems, 35:14274–14289, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.047061Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:7fd19fe9f706c1e0b7dfacb849a24d5281aa2db465283d11a77b61ecf24edc99","observation_id":"3d4c001b-8268-4ec6-8f9e-3962a1ddc3bf","resolution":{"observed_at":"2026-08-03T10:52:03.047061Z","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":"2026-08-03T10:52:03.052657Z","title":"Few-shot, now for real: Medi- cal VLMs adaptation without balanced sets or validation","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.052657Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:30ce83cad9710b1b29fbe48272e8d337704ac8665e32e192101f130f98392ca7","observation_id":"8188788d-01ed-42f9-ba0a-3d2bce990432","resolution":{"observed_at":"2026-08-03T10:52:03.052657Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1212.0402","last_updated":"2012-12-03T14:45:31Z","snapshot_observed_at":"2026-07-06T03:01:10.229407Z","submitted_at":"2012-12-03T14:45:31Z","title":"UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1212.0402","snapshot_observed_at":"2026-08-03T10:52:03.063757Z","title":"Ucf101: A dataset of 101 human actions classes from videos in the wild.ArXiv, abs/1212.0402, 2012","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.063757Z"},"links":{"cited_paper":"/paper/1212.0402","citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:c6eaed420d0353dc99443e10a6c1883a157cb86a1ecffdf7b024649a784096e4","observation_id":"6e5b4bef-147c-4abc-88bd-0130a1ecfc2c","resolution":{"observed_at":"2026-08-03T10:52:03.063757Z","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":"2026-08-03T10:52:03.069655Z","title":"An empirical study into what matters for calibrating vision-language models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.069655Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:2bf892a3bff71183473bfe370da5866be462ac4bd6cc61d620945a97d3466b2f","observation_id":"8dd32646-e4ae-45fb-8340-da0349f7a204","resolution":{"observed_at":"2026-08-03T10:52:03.069655Z","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":"2026-08-03T10:52:03.082824Z","title":"Learning robust global representations by penalizing local predictive power","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.082824Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:aed5ff8e9297848d77922fcfbb6383f345159ae6ff9396d9ba57686faf891baf","observation_id":"df57c9ec-21f3-42eb-8e61-fe62204f38ec","resolution":{"observed_at":"2026-08-03T10:52:03.082824Z","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":"2026-08-03T10:52:03.088904Z","title":"Open-vocabulary calibration for fine-tuned CLIP","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.088904Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:ab9295961ed10334cb590173663c4dafa4fd26ab9433e492e581d985920acbf9","observation_id":"4c204992-fda6-4351-906b-127a97ef4e19","resolution":{"observed_at":"2026-08-03T10:52:03.088904Z","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":"2026-08-03T10:52:03.094112Z","title":"Sun database: Large-scale scene 10 recognition from abbey to zoo","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.094112Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:e23866c15a739ed6942d70f8572516ad9caac466e1dfc92168b33b2ddef275d9","observation_id":"23c52554-ff37-49f0-8f64-17bb55adfb84","resolution":{"observed_at":"2026-08-03T10:52:03.094112Z","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":"2026-08-03T10:52:03.100591Z","title":"Demysti- fying CLIP data","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.100591Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:2ab988f9060d9dd2433fbfc4676d39dfe160de910b965fe35f0e3d3a7230c42e","observation_id":"87fe0415-628f-4659-89e4-e38998a9b1e8","resolution":{"observed_at":"2026-08-03T10:52:03.100591Z","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":"2026-08-03T10:52:03.105431Z","title":"Beyond probability partitions: Calibrating neural networks with semantic aware grouping.Advances in Neural Information Processing Systems, 36:58448–58460,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.105431Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:ec1dd8c0ca2e732bd49af8f62f1fafae98a4a9d0107c4774d5d450326bc9907c","observation_id":"5ace1d15-af96-40bd-ae95-466f67b7d6c0","resolution":{"observed_at":"2026-08-03T10:52:03.105431Z","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":"2026-08-03T10:52:03.110974Z","title":"Hasegawa-Johnson, Yingzhen Li, and Chang D","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.110974Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:ee4be5e34b44f5b6d7cb59b732d57ba64462aa570d1cfb9c95cc2ca0232aebf0","observation_id":"aa8afaec-3e22-4df4-b722-abdf03a30f42","resolution":{"observed_at":"2026-08-03T10:52:03.110974Z","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":"2026-08-03T10:52:03.116538Z","title":"Cutmix: Regular- ization strategy to train strong classifiers with localizable fea- tures","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.116538Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:a3a85da64ee6c9e01dadffca734b4102b32157d892ccaa10e22691c3a92a5ba8","observation_id":"6896931d-c33d-4b08-8237-34f3cf2f3aa4","resolution":{"observed_at":"2026-08-03T10:52:03.116538Z","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":"2026-08-03T10:52:03.122156Z","title":"Dauphin, and David Lopez-Paz","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.122156Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:0383d1875682da36c2d9907374109d6193f7389f2be2ed31bc9208b305b58a6d","observation_id":"9a171071-70d7-461e-943a-545bd1d78ed7","resolution":{"observed_at":"2026-08-03T10:52:03.122156Z","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":"2026-08-03T10:52:03.127331Z","title":"Mix-n-match: Ensemble and compositional methods for uncertainty calibration in deep learning","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.127331Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:9828ae496a22978014363510de525f3370c135af3630ff36071baf2d03bd52ab","observation_id":"05978e98-4aad-42c7-a878-b4d93aca1c6b","resolution":{"observed_at":"2026-08-03T10:52:03.127331Z","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":"2026-08-03T10:52:03.132563Z","title":"Conditional prompt learning for vision-language models","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.132563Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:7f1336c913b18404db289dd3a79c35c6bcaa7321905392a1726ffcb6e12490e7","observation_id":"731ee871-f496-45fd-9911-98a3f9a9e62f","resolution":{"observed_at":"2026-08-03T10:52:03.132563Z","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":"2026-08-03T10:52:03.139813Z","title":"Learning to prompt for vision-language models.Inter- national Journal of Computer Vision (IJCV), 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:03.139813Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:c68d95d47ed6633063b68442f96164e78f63292352706941d310b0956cf2c4e3","observation_id":"3e9b26f4-0745-4ade-ac7b-82b0f3e8266f","resolution":{"observed_at":"2026-08-03T10:52:03.139813Z","resolver_source":null,"status":"malformed_identifier"},"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":"2026-08-03T10:52:02.789054Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning","version":1},"reference_index":2022,"source":"pdf_text","source_observed_at":"2026-08-03T10:52:02.789054Z"},"links":{"citing_paper":"/paper/2601.08617"},"observation_digest":"sha256:808f88b14d88f8bbd5bfc6e1792b2c7d13cb4e65cc771c1be2c8bdd10e420f26","observation_id":"e21361fd-2f78-45fe-862f-a87546526746","resolution":{"observed_at":"2026-08-03T10:52:02.789054Z","resolver_source":null,"status":"parse_uncertain"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2601.08617","last_updated":"2026-01-13T15:00:03Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-10T12:46:22.329950Z","submitted_at":"2026-01-13T15:00:03Z","title":"SoC: Semantic Orthogonal Calibration for Test-Time Prompt Tuning"},"reference_resolution":{"displayed":57,"state_counts":{"malformed_identifier":1,"metadata_mismatch":0,"parse_uncertain":1,"unresolved":55,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":57},"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 57 of 57 outbound references and 0 inbound Pith citation observations for arXiv:2601.08617."}