{"as_of":"2026-08-08T11:04:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:f7be115e19468c4d07d4b153b5f09fd1502cb2720ac163b2129dd23cb61c57ee","coverage":[{"denominator":0,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":7,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":7,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-08T06:32:00.761636+00:00","state":"measured"},{"denominator":7,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":7,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-07T12:40:21.446432Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-05-20T07:18:07.001231Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2305.19249","last_updated":"2023-05-30T17:35:31Z","snapshot_observed_at":"2026-07-06T15:35:33.952097Z","submitted_at":"2023-05-30T17:35:31Z","title":"Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.19249","snapshot_observed_at":"2026-08-07T12:40:21.446432Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.24147","last_updated":"2025-05-30T02:39:37Z","snapshot_observed_at":"2026-08-08T02:48:29.524773Z","submitted_at":"2025-05-30T02:39:37Z","title":"Rationales Are Not Silver Bullets: Measuring the Impact of Rationales on Model Performance and Reliability","version":1},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-08-07T12:40:21.446432Z"},"links":{"cited_paper":"/paper/2305.19249","citing_paper":"/paper/2505.24147"},"observation_digest":"sha256:1a1443ede10c5d3d3e31f0676d3df453e583fa3f9eb988c9a5fe02248bb2575b","observation_id":"6a428f17-441c-4495-b920-10589e7e418f","resolution":{"observed_at":"2026-08-07T12:40:21.446432Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.19249","last_updated":"2023-05-30T17:35:31Z","snapshot_observed_at":"2026-07-06T15:35:33.952097Z","submitted_at":"2023-05-30T17:35:31Z","title":"Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.19249","snapshot_observed_at":"2026-08-07T05:38:25.847025Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.07461","last_updated":"2025-06-09T06:10:04Z","snapshot_observed_at":"2026-08-07T22:27:02.755086Z","submitted_at":"2025-06-09T06:10:04Z","title":"From Calibration to Collaboration: LLM Uncertainty Quantification Should Be More Human-Centered","version":1},"reference_index":2022,"source":"pdf_text","source_observed_at":"2026-08-07T05:38:25.847025Z"},"links":{"cited_paper":"/paper/2305.19249","citing_paper":"/paper/2506.07461"},"observation_digest":"sha256:5e5ca7e1f5d87011eb1a1d6bcfa2db8f2ef88113920e3c899910ec4937481f75","observation_id":"b851a32c-b487-4a7b-b4ff-a4f29c10ff30","resolution":{"observed_at":"2026-08-07T05:38:25.847025Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.19249","last_updated":"2023-05-30T17:35:31Z","snapshot_observed_at":"2026-07-06T15:35:33.952097Z","submitted_at":"2023-05-30T17:35:31Z","title":"Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.19249","snapshot_observed_at":"2026-08-06T22:44:44.002754Z","title":"Preserving pre-trained features helps calibrate fine-tuned language models.ArXiv preprint, abs/2305.19249, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.20917","last_updated":"2025-06-26T00:49:35Z","snapshot_observed_at":"2026-08-07T05:17:23.851973Z","submitted_at":"2025-06-26T00:49:35Z","title":"Optimising Language Models for Downstream Tasks: A Post-Training Perspective","version":1},"reference_index":79,"source":"pdf_text","source_observed_at":"2026-08-06T22:44:44.002754Z"},"links":{"cited_paper":"/paper/2305.19249","citing_paper":"/paper/2506.20917"},"observation_digest":"sha256:25329600906207581a5089ed73928ab0c42305cbf61ca04aaea7b80c0f69a7a0","observation_id":"508e78f9-fa73-48d0-9ba2-bb3920c27b89","resolution":{"observed_at":"2026-08-06T22:44:44.002754Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.19249","last_updated":"2023-05-30T17:35:31Z","snapshot_observed_at":"2026-07-06T15:35:33.952097Z","submitted_at":"2023-05-30T17:35:31Z","title":"Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.19249","snapshot_observed_at":"2026-08-07T10:18:52.749933Z","title":"Preserving pre-trained features helps cal- ibrate fine-tuned language models,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.02870","last_updated":"2025-06-06T10:50:08Z","snapshot_observed_at":"2026-08-08T01:09:03.836574Z","submitted_at":"2025-06-06T10:50:08Z","title":"Loki's Dance of Illusions: A Comprehensive Survey of Hallucination in Large Language Models","version":1},"reference_index":72,"source":"pdf_text","source_observed_at":"2026-08-07T10:18:52.749933Z"},"links":{"cited_paper":"/paper/2305.19249","citing_paper":"/paper/2507.02870"},"observation_digest":"sha256:7c93244a001c5fc5a588f584e6df2f0247b248023e370bcdb186e3b450e4bb43","observation_id":"490c1a7e-70cc-4aae-bfb3-da5932207987","resolution":{"observed_at":"2026-08-07T10:18:52.749933Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.19249","last_updated":"2023-05-30T17:35:31Z","snapshot_observed_at":"2026-07-06T15:35:33.952097Z","submitted_at":"2023-05-30T17:35:31Z","title":"Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":"2305.19249","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.19249","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Preserving pre-trained features helps calibrate fine-tuned language models","venue":null,"work_id":"5b9114b4-50e7-4063-99d3-2951b6f31a88","year":2023},"citing_paper":{"arxiv_id":"2605.20005","last_updated":"2026-05-19T15:36:52Z","snapshot_observed_at":"2026-07-06T23:30:39.512029Z","submitted_at":"2026-05-19T15:36:52Z","title":"Fine-Tuning Without Forgetting via Loss-Adaptive Learning Rates","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-05-20T07:14:59.396900Z"},"links":{"cited_paper":"/paper/2305.19249","citing_paper":"/paper/2605.20005"},"observation_digest":"sha256:65f16723c06d0cf9a1c8cd975bcbce4baa4fa6e3acec44f80d1c24cbead8a086","observation_id":"0a2ad652-dbb3-4615-91d5-78c3c7c368f6","resolution":{"observed_at":"2026-05-20T07:18:07.003009Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"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"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.19249","last_updated":"2023-05-30T17:35:31Z","snapshot_observed_at":"2026-07-06T15:35:33.952097Z","submitted_at":"2023-05-30T17:35:31Z","title":"Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.19249","snapshot_observed_at":"2026-07-14T07:42:06.462698Z","title":"arXiv preprint arXiv:2305.19249 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.11007","last_updated":"2026-07-28T06:21:45Z","snapshot_observed_at":"2026-08-02T18:37:42.251838Z","submitted_at":"2026-07-13T02:11:27Z","title":"TabPFN beyond Tabular Data: Calibration and Accuracy on Multimodal Embeddings","version":1},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-07-14T07:42:06.462698Z"},"links":{"cited_paper":"/paper/2305.19249","citing_paper":"/paper/2607.11007"},"observation_digest":"sha256:3d1ed6caa9473376bfb5fe5fec975b93b1eb527ddc1be86bb47aa024e76c3615","observation_id":"9363f261-8615-4810-b1a8-b28d4011c611","resolution":{"observed_at":"2026-07-14T07:42:06.462698Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.19249","last_updated":"2023-05-30T17:35:31Z","snapshot_observed_at":"2026-07-06T15:35:33.952097Z","submitted_at":"2023-05-30T17:35:31Z","title":"Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.19249","snapshot_observed_at":"2026-08-02T07:07:57.295329Z","title":"arXiv preprint arXiv:2305.19249 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.11007","last_updated":"2026-07-28T06:21:45Z","snapshot_observed_at":"2026-08-02T18:37:42.251838Z","submitted_at":"2026-07-13T02:11:27Z","title":"TabPFN beyond Tabular Data: Calibration and Accuracy on Multimodal Embeddings","version":3},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-08-02T07:07:57.295329Z"},"links":{"cited_paper":"/paper/2305.19249","citing_paper":"/paper/2607.11007"},"observation_digest":"sha256:6993f41639a9d6d72ecff3d985e7044c114d174e055c5e043bd4717ce994861d","observation_id":"6174ec00-a6bd-4ed3-a318-c8d8c68cea11","resolution":{"observed_at":"2026-08-02T07:07:57.295329Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2305.19249/citation-record","integrity":"/paper/2305.19249/integrity","json":"/paper/2305.19249/citation-record.json","paper":"/paper/2305.19249"},"outbound":[],"paper":{"arxiv_id":"2305.19249","last_updated":"2023-05-30T17:35:31Z","latest_version":1,"primary_category":"cs.CL","snapshot_observed_at":"2026-07-06T15:35:33.952097Z","submitted_at":"2023-05-30T17:35:31Z","title":"Preserving Pre-trained Features Helps Calibrate Fine-tuned Language Models"},"reference_resolution":{"displayed":0,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":0,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":0},"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 0 of 0 outbound references and 7 inbound Pith citation observations for arXiv:2305.19249."}