{"as_of":"2026-08-05T13:43:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:a446bde6a1866463295e8a72722c09b22990e74de288dc4ab16b9eac40ddd9b3","coverage":[{"denominator":24,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":24,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-04T10:49:03.893003Z","state":"measured"},{"denominator":24,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":24,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-05T06:32:48.257954+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/2510.08734/citation-record","integrity":"/paper/2510.08734/integrity","json":"/paper/2510.08734/citation-record.json","paper":"/paper/2510.08734"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"2507.16003","last_updated":"2026-06-02T08:52:07Z","snapshot_observed_at":"2026-07-06T22:00:38.257294Z","submitted_at":"2025-07-21T18:44:35Z","title":"Learning without training: The implicit dynamics of in-context learning","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2507.16003","snapshot_observed_at":"2026-08-04T10:49:01.294460Z","title":"Learning without training: The implicit dynamics of in-context learning.arXiv preprint arXiv:2507.16003, July 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:01.294460Z"},"links":{"cited_paper":"/paper/2507.16003","citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:044fe4dee792385f8a5404bba11cabeb0671acce09ddc3e4d17822a06950dc1a","observation_id":"a56a102c-e363-4f25-9aab-45b14980081d","resolution":{"observed_at":"2026-08-04T10:49:01.294460Z","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":"2026-08-04T10:49:01.348027Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:01.348027Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:723244e4d91e0219c738c2955cd55e189f4c7670197c5162248b13e75aad06ab","observation_id":"426404f5-7477-42d6-aa73-989fb7fa6694","resolution":{"observed_at":"2026-08-04T10:49:01.348027Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2306.03341","last_updated":"2024-06-26T14:11:53Z","snapshot_observed_at":"2026-07-06T15:38:58.578200Z","submitted_at":"2023-06-06T01:26:53Z","title":"Inference-Time Intervention: Eliciting Truthful Answers from a Language Model","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2306.03341","snapshot_observed_at":"2026-08-04T10:49:01.441654Z","title":"Inference- time intervention: Eliciting truthful answers from a language model.arXiv preprint arXiv:2306.03341, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:01.441654Z"},"links":{"cited_paper":"/paper/2306.03341","citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:55bd0311ac9923b1bf960a0d6f6164df0ae3a1675554322f642cea56772b6d84","observation_id":"b4cd9f52-6fdc-4372-840e-25b89e9e7c09","resolution":{"observed_at":"2026-08-04T10:49:01.441654Z","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":"2026-08-04T10:49:01.578153Z","title":"Steering language models with activation engineering, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:01.578153Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:87d6787a5b89a288f9ab3e02fcc56eb59dfd5663b2d059018d2282d68367f653","observation_id":"7940ec54-2048-4b59-ad0e-e5880f325092","resolution":{"observed_at":"2026-08-04T10:49:01.578153Z","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":"2026-08-04T10:49:01.745268Z","title":"Function vectors in large language models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:01.745268Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:2c9d9774e4aba6e91b0672a0ec6fe2e778b5d383e49e896329f591771086b80d","observation_id":"7de65b63-e3c3-4b33-8882-1e8d5246a63f","resolution":{"observed_at":"2026-08-04T10:49:01.745268Z","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":"2026-08-04T10:49:01.897672Z","title":"Locating and editing factual associations in GPT","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:01.897672Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:8dfab71af18318b28aa9f2182280c3fd4140dd04cb8f6cd909ec7411757e4f2c","observation_id":"dde73895-17d7-442a-a9a7-1a0cc00276b3","resolution":{"observed_at":"2026-08-04T10:49:01.897672Z","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":"2026-08-04T10:49:02.042021Z","title":"Fast model editing at scale","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:02.042021Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:b0a75e36758112af7652de67eb28418fa7650bbf6595cb3810d07eb3844460e3","observation_id":"9dc0c8a5-1004-4ece-8308-ae8bce0888c4","resolution":{"observed_at":"2026-08-04T10:49:02.042021Z","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":"2026-08-04T10:49:02.215814Z","title":"Byun, Zifan Wang, Alex Mallen, Steven Basart, Sanmi Koyejo, Dawn Song, Matt Fredrikson, J","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:02.215814Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:ef134492e7573e2020e60cf5f5b6b473d762ba03a3e8d4d07f3c924c943ad759","observation_id":"3f752a1e-0956-4dbc-be13-cc04cf3456ba","resolution":{"observed_at":"2026-08-04T10:49:02.215814Z","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":"2026-08-04T10:49:02.391531Z","title":"A unified understanding and evaluation of steering methods.ArXiv, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:02.391531Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:d41f1cdbf2bc32db975fe18413ca6759d630cfe41ebd8fa9675004ba91c2fe7f","observation_id":"9b8dd970-7f31-4663-98b6-d2e3de118eb6","resolution":{"observed_at":"2026-08-04T10:49:02.391531Z","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":"2026-08-04T10:49:02.565274Z","title":"In-context learning creates task vectors","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:02.565274Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:8873400410884dec2042ebbc2cbde08cdf47b74585b9c8673e54727caecf2dd0","observation_id":"185ca4c5-3d1f-4173-9d18-0480aba1bd38","resolution":{"observed_at":"2026-08-04T10:49:02.565274Z","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":"2026-08-04T10:49:02.682353Z","title":"Analysing the generalisation and reliability of steering vectors","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:02.682353Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:334261ec7a13539a0b570f48f19f8ab171c2a56b31914b1dbc2175fc002c6096","observation_id":"d292a625-39ed-480e-ad22-5e7eb3694f02","resolution":{"observed_at":"2026-08-04T10:49:02.682353Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.17245","last_updated":"2024-10-22T17:59:39Z","snapshot_observed_at":"2026-07-06T19:38:00.663759Z","submitted_at":"2024-10-22T17:59:39Z","title":"Towards Reliable Evaluation of Behavior Steering Interventions in LLMs","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.17245","snapshot_observed_at":"2026-08-04T10:49:02.853156Z","title":"Towards reliable evaluation of behavior steering interventions in llms.ArXiv, abs/2410.17245, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:02.853156Z"},"links":{"cited_paper":"/paper/2410.17245","citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:f5cf56ef3ec4dbff8ce00ac8e5a6a129416e51a2a68d29df24d1a7ad6cac8890","observation_id":"b2781232-74a0-46df-9089-ef4bbb67230f","resolution":{"observed_at":"2026-08-04T10:49:02.853156Z","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":"2026-08-04T10:49:02.947155Z","title":"Comparing bottom-up and top-down steering approaches on in-context learning tasks.ArXiv, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:02.947155Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:f3cc49f1c02e66ef8ab19abeb89aa4034b6c7db0139a776485cc6d63a7d0bdfb","observation_id":"4a63b250-31ba-41ca-86a6-a93426f031a9","resolution":{"observed_at":"2026-08-04T10:49:02.947155Z","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":"2026-08-04T10:49:03.079086Z","title":"Task vectors in in-context learning: Emergence, formation, and benefit, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.079086Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:9cc8d5160767e4bdad0a9fefe697b713f21cdef8fc06e9036e4055ab30e17e30","observation_id":"d6dce485-a0b3-49f0-9ff5-afba5426b1c9","resolution":{"observed_at":"2026-08-04T10:49:03.079086Z","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":"2026-08-04T10:49:03.217802Z","title":"Transformer feed-forward layers are key-value memories, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.217802Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:b9097fe395156e1918f671a766405eab9ec48b7d907ae608411fadcb1de46293","observation_id":"cf025543-e954-4c4f-934f-590060de8385","resolution":{"observed_at":"2026-08-04T10:49:03.217802Z","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":"2026-08-04T10:49:03.336844Z","title":"Editing factual knowledge in language models, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.336844Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:10823587ae40023414da9cc83addb3c3a138d0a570e11ba2e8a22361af70bc84","observation_id":"4f332140-6f0c-405a-b533-a3617352cf00","resolution":{"observed_at":"2026-08-04T10:49:03.336844Z","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":"2026-08-04T10:49:03.422601Z","title":"Assessing the brittleness of safety alignment via pruning and low-rank modifications","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.422601Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:f642380c218f962ff089eb65fecefda1dbc90aa5691ea490f016c4a87c7ed0ff","observation_id":"e6c215ae-d0f6-4af7-932a-4e592580214c","resolution":{"observed_at":"2026-08-04T10:49:03.422601Z","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":"2026-08-04T10:49:03.507053Z","title":"Model editing as a robust and denoised variant of DPO: A case study on toxicity","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.507053Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:54a44262cfd5ae52babae1bf08d0401e5208231c0feba5bd26572483d1da972c","observation_id":"1e7b303a-11e5-4576-b2d4-4b3603562e6a","resolution":{"observed_at":"2026-08-04T10:49:03.507053Z","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":"2026-08-04T10:49:03.580933Z","title":"Editing models with task arithmetic","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.580933Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:0880ab29ba773e25bc7e347d0dde96d4165200311836301629e25c2557dfa178","observation_id":"440c5b6a-3a8f-4e95-92d2-e84a1fe31bbc","resolution":{"observed_at":"2026-08-04T10:49:03.580933Z","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":"2026-08-04T10:49:03.624189Z","title":"What can transformers learn in-context? a case study of simple function classes","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.624189Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:f8663b821a668cb52345a9bb1dea7ce9d15b12c0da7839014f811936eb77c9b4","observation_id":"3d2c7d07-81de-4afc-8b65-282310bd94d9","resolution":{"observed_at":"2026-08-04T10:49:03.624189Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.19786","last_updated":"2025-03-25T15:52:34Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2025-03-25T15:52:34Z","title":"Gemma 3 Technical Report","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.19786","snapshot_observed_at":"2026-08-04T10:49:03.689025Z","title":"Gemma 3 Technical Report.arXiv preprint arXiv:2503.19786, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.689025Z"},"links":{"cited_paper":"/paper/2503.19786","citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:7baf9838c2941b0d3a5ba3cc0ebc3a16cd9f5d50830a8a04e987980c610b0dea","observation_id":"954e001c-f001-4f15-a437-ddbcd6302fbe","resolution":{"observed_at":"2026-08-04T10:49:03.689025Z","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":"2026-08-04T10:49:03.776511Z","title":", yn is a basis of the space (which implies that n=d ), then the inverse takes the form Z= X i ωiωT i , where the vectors ωi are the rows of Y −1","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.776511Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:430db9591ffe063bad7424fb071719c4470a503bd11a6edc6d10b467cbabbcfa","observation_id":"bdbe1ac6-6bb3-4cdc-a9f6-240e93c78d5d","resolution":{"observed_at":"2026-08-04T10:49:03.776511Z","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":"2026-08-04T10:49:03.834442Z","title":", yn is an orthonormal basis (n=d) of the space Z −1 =I","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.834442Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:da525bfeff20bca8066c1e1188297a66dd84f53013ecfaea47672cf8a59de4e3","observation_id":"7f74b036-c9ed-4d17-90f7-c52554010969","resolution":{"observed_at":"2026-08-04T10:49:03.834442Z","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":"2026-08-04T10:49:03.893003Z","title":", yn are vectors independently sampled from a spherical distribution, then for nlarge enough Z −1 = 1 σ2n I, whereσ 2 is the distribution variance","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights","version":3},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-04T10:49:03.893003Z"},"links":{"citing_paper":"/paper/2510.08734"},"observation_digest":"sha256:b63b7dee66bfc0a58d426c23634b3b70ff362f1cd0fa71d882a7ddd8ef5b12dc","observation_id":"e8467624-2cba-4aad-ac6b-1ef08f6fcb37","resolution":{"observed_at":"2026-08-04T10:49:03.893003Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2510.08734","last_updated":"2026-06-03T12:25:25Z","latest_version":3,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-04T10:49:00.487013Z","submitted_at":"2025-10-09T18:40:39Z","title":"Transmuting prompts into weights"},"reference_resolution":{"displayed":24,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":24,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":24},"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":"2026-08-05T06:32:48.257954+00:00","source":"crossref"},{"observed_at":"2026-08-05T06:32:44.755628+00:00","source":"retraction_watch"}],"thesis":"As of 5 August 2026, this Paper Citation Record lists 24 of 24 outbound references and 0 inbound Pith citation observations for arXiv:2510.08734."}