{"as_of":"2026-08-22T06:34:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:cb78f32e9067eca8a317429f532bb090a0691d128ffdf280478242e41ce410d1","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":69,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":69,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-22T06:32:14.747728+00:00","state":"measured"},{"denominator":69,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":69,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-16T10:14:32.340920Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-07-04T18:40:03.102880Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2303.11366","last_updated":"2023-10-10T05:21:45Z","snapshot_observed_at":"2026-08-17T19:15:32.679249Z","submitted_at":"2023-03-20T18:08:50Z","title":"Reflexion: Language Agents with Verbal Reinforcement Learning","version":4},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-05-10T13:51:41.915864Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2303.11366"},"observation_digest":"sha256:0913e99e3d3a9658e43b29264305447b6066d4e9d04d9610e8ebacdb661e84ea","observation_id":"c21e125a-3242-4b21-a107-19862b44f6e0","resolution":{"observed_at":"2026-05-10T13:51:41.963062Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2309.03409","last_updated":"2024-04-15T07:50:32Z","snapshot_observed_at":"2026-08-15T09:21:11.223719Z","submitted_at":"2023-09-07T00:07:15Z","title":"Large Language Models as Optimizers","version":3},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-05-15T00:04:31.212102Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2309.03409"},"observation_digest":"sha256:e84ffe8561d642c5fc67c78741bebf10e00775f162ee8bf0026c578d156d9ce4","observation_id":"248d197e-5664-46a0-b283-9a17060d7292","resolution":{"observed_at":"2026-05-15T00:04:31.319892Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2309.08532","last_updated":"2025-05-01T11:56:52Z","snapshot_observed_at":"2026-08-21T11:33:21.306052Z","submitted_at":"2023-09-15T16:50:09Z","title":"EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizers","version":3},"reference_index":124,"source":"arxiv_source","source_observed_at":"2026-05-16T06:11:49.475825Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2309.08532"},"observation_digest":"sha256:d207cc33ff43d34d7c13a915af8a1fed1f0e460b6767dafac153215a93cdc95a","observation_id":"becd4a9f-14a8-495a-99a4-ab0364ab9764","resolution":{"observed_at":"2026-05-16T06:11:49.632758Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2309.16797","last_updated":"2023-09-28T19:01:07Z","snapshot_observed_at":"2026-08-14T11:37:32.461743Z","submitted_at":"2023-09-28T19:01:07Z","title":"Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution","version":1},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-05-16T08:12:30.984870Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2309.16797"},"observation_digest":"sha256:39eb3572166d42d064938995d7cf0b6c27fa30127cc1cd2b2aa648abed3b974d","observation_id":"213284a3-377b-4f83-b988-8249236fa8cc","resolution":{"observed_at":"2026-05-16T08:12:35.372889Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2310.03714","last_updated":"2023-10-05T17:37:25Z","snapshot_observed_at":"2026-08-04T05:21:05.846165Z","submitted_at":"2023-10-05T17:37:25Z","title":"DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines","version":1},"reference_index":42,"source":"arxiv_source","source_observed_at":"2026-05-11T18:57:46.756656Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2310.03714"},"observation_digest":"sha256:10ffc06882db294b2ecd8f0a006ca507ec8556ae5054ef79e1954986b3ec06dd","observation_id":"48a488a9-4799-4f18-b2c6-34b41ca9b20b","resolution":{"observed_at":"2026-05-11T18:57:46.984559Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2310.11324","last_updated":"2024-07-01T22:28:01Z","snapshot_observed_at":"2026-08-12T15:05:30.306144Z","submitted_at":"2023-10-17T15:03:30Z","title":"Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting","version":2},"reference_index":53,"source":"arxiv_source","source_observed_at":"2026-05-17T01:59:51.326493Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2310.11324"},"observation_digest":"sha256:51fa5cbed58c3536a0fd956073cbecdf3995d17e647a99fd3ef995626a8997f1","observation_id":"5ac2f298-efaf-462f-9ad2-c7d11eb7e95e","resolution":{"observed_at":"2026-05-17T01:59:51.390802Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-11T23:49:47.777789Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2412.02173","last_updated":"2024-12-03T05:05:13Z","snapshot_observed_at":"2026-08-17T22:46:46.365555Z","submitted_at":"2024-12-03T05:05:13Z","title":"Keeping Experts in the Loop: Expert-Guided Optimization for Clinical Data Classification using Large Language Models","version":1},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-08-11T23:49:47.777789Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2412.02173"},"observation_digest":"sha256:d36f06c03ed48190dbd8a324656c57d3c3831123d458148212c5fa7bc5d9cb56","observation_id":"42853ad6-8802-4dd5-a3bb-1204e374f4e0","resolution":{"observed_at":"2026-08-11T23:49:47.777789Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-11T22:52:15.153246Z","title":"gradient descent","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.03092","last_updated":"2025-06-18T04:06:03Z","snapshot_observed_at":"2026-08-17T07:11:23.935721Z","submitted_at":"2024-12-04T07:44:35Z","title":"REVOLVE: Optimizing AI Systems by Tracking Response Evolution in Textual Optimization","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-11T22:52:15.153246Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2412.03092"},"observation_digest":"sha256:8f85718866cd7fd990f75ae14f79cb517dfd1580a405ea35cc81462021fb136e","observation_id":"ff0f7aca-7e9a-48b8-b17f-c48348e5c2f2","resolution":{"observed_at":"2026-08-11T22:52:15.153246Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-11T16:52:28.858246Z","title":"Automatic prompt optimization with” gradient descent” and beam search","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.09722","last_updated":"2025-04-07T15:39:16Z","snapshot_observed_at":"2026-08-16T17:27:32.658393Z","submitted_at":"2024-12-12T20:59:43Z","title":"GReaTer: Gradients over Reasoning Makes Smaller Language Models Strong Prompt Optimizers","version":2},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-11T16:52:28.858246Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2412.09722"},"observation_digest":"sha256:5849691f5dbc43c35f5703a1651c4c5ff95b4212a56b8effe43eb8edcc0400a6","observation_id":"58ee331e-028c-4880-b473-3a0245e8a01d","resolution":{"observed_at":"2026-08-11T16:52:28.858246Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-11T12:08:03.446562Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2412.14660","last_updated":"2024-12-25T06:05:36Z","snapshot_observed_at":"2026-08-18T13:06:26.993824Z","submitted_at":"2024-12-19T09:10:07Z","title":"Unveiling Uncertainty: A Deep Dive into Calibration and Performance of Multimodal Large Language Models","version":2},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-08-11T12:08:03.446562Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2412.14660"},"observation_digest":"sha256:9e68d0769a16b234c21054452a078e3f144cffabb57225983263e6d9aeb344cd","observation_id":"15a976c4-3c5f-4b22-944a-7a56d05c7142","resolution":{"observed_at":"2026-08-11T12:08:03.446562Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-11T13:27:32.007712Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2412.15274","last_updated":"2024-12-17T18:35:04Z","snapshot_observed_at":"2026-08-15T08:20:34.221577Z","submitted_at":"2024-12-17T18:35:04Z","title":"Memory-Augmented Agent Training for Business Document Understanding","version":1},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-08-11T13:27:32.007712Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2412.15274"},"observation_digest":"sha256:320be22e469d959cfc97444584efb5970026c1879e2adf06e3dd323ebf925226","observation_id":"6ec70e26-31ac-470f-af77-f1c385d5afd8","resolution":{"observed_at":"2026-08-11T13:27:32.007712Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-11T00:08:16.262491Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2412.19684","last_updated":"2025-02-17T15:32:50Z","snapshot_observed_at":"2026-08-18T09:40:01.388208Z","submitted_at":"2024-12-27T15:21:17Z","title":"Boosting Private Domain Understanding of Efficient MLLMs: A Tuning-free, Adaptive, Universal Prompt Optimization Framework","version":2},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-11T00:08:16.262491Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2412.19684"},"observation_digest":"sha256:1374f142bb84b19ee091f97da37c45bf92f7131f5f13055e701869d805e0730d","observation_id":"370d08af-cab4-4270-8fc0-00d151c0797d","resolution":{"observed_at":"2026-08-11T00:08:16.262491Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-10T17:14:06.503198Z","title":"Automatic prompt optimization with","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.12521","last_updated":"2025-01-21T22:24:03Z","snapshot_observed_at":"2026-08-13T10:08:12.836488Z","submitted_at":"2025-01-21T22:24:03Z","title":"An Empirically-grounded tool for Automatic Prompt Linting and Repair: A Case Study on Bias, Vulnerability, and Optimization in Developer Prompts","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-10T17:14:06.503198Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2501.12521"},"observation_digest":"sha256:44686a376c545d99d575efc6c8bb0a4af3e0174f1b9acb6efc667df72970ab57","observation_id":"25a67637-65b7-4a28-8045-d49684bf27b2","resolution":{"observed_at":"2026-08-10T17:14:06.503198Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-10T11:33:24.646115Z","title":"Automatic prompt optimization with” gradient descent” and beam search","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.16673","last_updated":"2025-01-30T16:40:12Z","snapshot_observed_at":"2026-08-19T00:43:00.993791Z","submitted_at":"2025-01-28T03:18:48Z","title":"LLM-AutoDiff: Auto-Differentiate Any LLM Workflow","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-10T11:33:24.646115Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2501.16673"},"observation_digest":"sha256:790ecf9715651ec8920a2c1325ae9a1860d1c888369d1e084ffdee9faca528d0","observation_id":"72977d4d-6982-49b8-b0ef-fe6e6f1df630","resolution":{"observed_at":"2026-08-10T11:33:24.646115Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-08T21:45:45.383025Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2502.04728","last_updated":"2025-05-08T13:42:18Z","snapshot_observed_at":"2026-08-15T02:02:52.968189Z","submitted_at":"2025-02-07T07:52:25Z","title":"Generating Symbolic World Models via Test-time Scaling of Large Language Models","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-08T21:45:45.383025Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2502.04728"},"observation_digest":"sha256:3774c7eb33f03da5460c46678d12eb4ebf1c9b94bffb7fc422a8f3d37285c1c1","observation_id":"ed5725ed-d274-40e8-89c1-761ee62b7939","resolution":{"observed_at":"2026-08-08T21:45:45.383025Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-08T12:22:43.278203Z","title":"arXiv preprint arXiv:2305.03495 (2023)","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2502.07556","last_updated":"2025-02-11T13:48:11Z","snapshot_observed_at":"2026-08-14T11:17:27.597317Z","submitted_at":"2025-02-11T13:48:11Z","title":"SketchFlex: Facilitating Spatial-Semantic Coherence in Text-to-Image Generation with Region-Based Sketches","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-08T12:22:43.278203Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2502.07556"},"observation_digest":"sha256:4b16b791214cf8beed88d7c5edb37d2b573883d72f06f7e505cb3755f48c811c","observation_id":"b0035b54-31fd-4826-8f7f-8f36abe1e12e","resolution":{"observed_at":"2026-08-08T12:22:43.278203Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-16T10:14:32.340920Z","title":"Gradient Descent","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2504.18722","last_updated":"2025-04-25T22:20:04Z","snapshot_observed_at":"2026-08-20T17:36:58.373544Z","submitted_at":"2025-04-25T22:20:04Z","title":"MODP: Multi Objective Directional Prompting","version":1},"reference_index":2023,"source":"pdf_text","source_observed_at":"2026-08-16T10:14:32.340920Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2504.18722"},"observation_digest":"sha256:a5baf9f47f6c4bafb5d5f487fdba60c349500eae2687b4b07b721f4233df3538","observation_id":"9cb9fe6e-734b-47b7-ae38-2066156d5ec2","resolution":{"observed_at":"2026-08-16T10:14:32.340920Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-16T05:35:33.343962Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.20355","last_updated":"2025-04-29T01:45:47Z","snapshot_observed_at":"2026-08-19T13:14:43.947510Z","submitted_at":"2025-04-29T01:45:47Z","title":"Local Prompt Optimization","version":1},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-08-16T05:35:33.343962Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2504.20355"},"observation_digest":"sha256:b435a2d31e2e15365b2978a1f9a42ea7211eb64bf277f1cb46b939e3125f9141","observation_id":"fcc4b4eb-1059-4a06-adf4-331c566e6781","resolution":{"observed_at":"2026-08-16T05:35:33.343962Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-15T23:32:07.515621Z","title":"Pryzant, D","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.04480","last_updated":"2025-05-07T14:51:43Z","snapshot_observed_at":"2026-08-18T03:56:15.488582Z","submitted_at":"2025-05-07T14:51:43Z","title":"TrajEvo: Designing Trajectory Prediction Heuristics via LLM-driven Evolution","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-15T23:32:07.515621Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2505.04480"},"observation_digest":"sha256:a51f03471854accc5fac3d107ba03f9c2a641eea68fa9cb5a79b146bb9952091","observation_id":"dbae75c1-ee3e-4eb9-b514-2d082e80b1c6","resolution":{"observed_at":"2026-08-15T23:32:07.515621Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-15T22:00:35.734403Z","title":"gradient descent","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2505.08303","last_updated":"2025-05-13T07:26:56Z","snapshot_observed_at":"2026-08-17T12:35:07.596926Z","submitted_at":"2025-05-13T07:26:56Z","title":"Evaluating the Effectiveness of Black-Box Prompt Optimization as the Scale of LLMs Continues to Grow","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-15T22:00:35.734403Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2505.08303"},"observation_digest":"sha256:edeaa140dd89480271d314e68524e3bee9cf1ebfacbe03ef9ccd2d3ac13bb711","observation_id":"06b76652-7bd8-4a69-9cb7-6fb1955cdbc1","resolution":{"observed_at":"2026-08-15T22:00:35.734403Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-15T21:45:35.101965Z","title":"”Automatic prompt optimization with” gradient descent” and beam search.” arXiv preprint arXiv:2305.03495 (2023)","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.09029","last_updated":"2025-05-13T23:56:12Z","snapshot_observed_at":"2026-08-20T11:05:33.705067Z","submitted_at":"2025-05-13T23:56:12Z","title":"Monte Carlo Beam Search for Actor-Critic Reinforcement Learning in Continuous Control","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-15T21:45:35.101965Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2505.09029"},"observation_digest":"sha256:bfb1648626ba18acbffba17c10f07ab81bff58c210aa33f496b48bc54ceec68d","observation_id":"866b4b9a-3db6-4e01-aa06-335aa0eef24b","resolution":{"observed_at":"2026-08-15T21:45:35.101965Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-15T21:08:46.207594Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.10736","last_updated":"2025-08-18T18:02:26Z","snapshot_observed_at":"2026-08-19T18:50:27.154970Z","submitted_at":"2025-05-15T22:41:30Z","title":"Model Performance-Guided Evaluation Data Selection for Effective Prompt Optimization","version":3},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-15T21:08:46.207594Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2505.10736"},"observation_digest":"sha256:9c12ee76af2f01db7ef0b59f6ae5253509998fb0bbaab8ff61f5be9415047198","observation_id":"68c6f579-4b73-4c96-a89c-7fa29cd3b437","resolution":{"observed_at":"2026-08-15T21:08:46.207594Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-07T15:08:25.999511Z","title":"Automatic prompt optimization with","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.17137","last_updated":"2025-09-03T03:28:34Z","snapshot_observed_at":"2026-08-20T11:14:36.438447Z","submitted_at":"2025-05-22T05:40:12Z","title":"Cog-TiPRO: Iterative Prompt Refinement with LLMs to Detect Cognitive Decline via Longitudinal Voice Assistant Commands","version":4},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-07T15:08:25.999511Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2505.17137"},"observation_digest":"sha256:a7581acd11183ec15130c80c81a0846977e16b08e382b99c98943f14e583ff1f","observation_id":"ed60016a-60bb-4880-970b-71b09668513e","resolution":{"observed_at":"2026-08-07T15:08:25.999511Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-07T13:19:32.679972Z","title":"Automatic prompt optimization with","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.22112","last_updated":"2025-05-28T08:40:55Z","snapshot_observed_at":"2026-08-19T04:49:55.590336Z","submitted_at":"2025-05-28T08:40:55Z","title":"Visual Large Language Models Exhibit Human-Level Cognitive Flexibility in the Wisconsin Card Sorting Test","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-07T13:19:32.679972Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2505.22112"},"observation_digest":"sha256:3ee18ca86ff0ab1a26790de7f5cbcd84ddd5d11333c44bddac934af70206aa52","observation_id":"67ae6db2-43e4-42a9-a8a1-91266ea4ffdc","resolution":{"observed_at":"2026-08-07T13:19:32.679972Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-07T12:15:39.830669Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.00178","last_updated":"2025-07-22T18:01:11Z","snapshot_observed_at":"2026-08-07T23:16:23.624156Z","submitted_at":"2025-05-30T19:33:41Z","title":"Tournament of Prompts: Evolving LLM Instructions Through Structured Debates and Elo Ratings","version":2},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-07T12:15:39.830669Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2506.00178"},"observation_digest":"sha256:c96dfca57f24ee6a8023701dc93770a13149cbb3b8fec2258001483957431480","observation_id":"03b6f9bb-dec4-4d97-89a5-6a47e344f3db","resolution":{"observed_at":"2026-08-07T12:15:39.830669Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-07T12:05:56.352320Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.00794","last_updated":"2025-06-01T02:46:31Z","snapshot_observed_at":"2026-08-21T21:23:10.956041Z","submitted_at":"2025-06-01T02:46:31Z","title":"Predicting Empirical AI Research Outcomes with Language Models","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-07T12:05:56.352320Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2506.00794"},"observation_digest":"sha256:8c7cd6e49c84883418fa1094a9b396f6b2f5ee2ef238b75950af7ad121850ff3","observation_id":"0dde4865-f67f-48c4-b329-bb5f47e9ba75","resolution":{"observed_at":"2026-08-07T12:05:56.352320Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-07T13:24:34.774410Z","title":"Automatic prompt optimization with","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.02014","last_updated":"2025-05-28T02:22:11Z","snapshot_observed_at":"2026-08-19T14:52:39.606148Z","submitted_at":"2025-05-28T02:22:11Z","title":"Research on Driving Scenario Technology Based on Multimodal Large Lauguage Model Optimization","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-07T13:24:34.774410Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2506.02014"},"observation_digest":"sha256:c5fc7a7d6de6ff7715d5bd7bef560d2a9440b9b3de80f99e93c37bbaab23f58d","observation_id":"80a63cd2-c49e-4284-80c9-33203b0a57ce","resolution":{"observed_at":"2026-08-07T13:24:34.774410Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-07T10:19:34.725290Z","title":"gradient descent","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.06401","last_updated":"2025-06-06T02:40:42Z","snapshot_observed_at":"2026-08-15T11:54:44.539162Z","submitted_at":"2025-06-06T02:40:42Z","title":"Direct Behavior Optimization: Unlocking the Potential of Lightweight LLMs","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-07T10:19:34.725290Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2506.06401"},"observation_digest":"sha256:fd66b8b06bc2b70ba37ef75e57e0b025e61a4e0d6059c1ae078b0bc4b689505d","observation_id":"c1c5b9b8-fa9c-4d5b-8bfe-23dae174d260","resolution":{"observed_at":"2026-08-07T10:19:34.725290Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-07T05:10:39.559056Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.08669","last_updated":"2025-06-10T10:30:43Z","snapshot_observed_at":"2026-08-20T22:29:01.952768Z","submitted_at":"2025-06-10T10:30:43Z","title":"Enhancing Reasoning Capabilities of Small Language Models with Blueprints and Prompt Template Search","version":1},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-08-07T05:10:39.559056Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2506.08669"},"observation_digest":"sha256:8ff619c73a441294957d280c9529894836af06b7aebb2daffef53f3522f7b153","observation_id":"9817da71-1d6d-4c95-bca6-6f241c3dd680","resolution":{"observed_at":"2026-08-07T05:10:39.559056Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-15T19:07:06.877853Z","title":"arXiv preprint arXiv:2305.03495 (2023)","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.17700","last_updated":"2025-06-21T12:25:37Z","snapshot_observed_at":"2026-08-18T09:26:29.102965Z","submitted_at":"2025-06-21T12:25:37Z","title":"The Evolution of Natural Language Processing: How Prompt Optimization and Language Models are Shaping the Future","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-15T19:07:06.877853Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2506.17700"},"observation_digest":"sha256:cd77b0383ed525dc15d4eeeae4b4241b48bb47b6e6c5c9caf9d3c50b1254a7a7","observation_id":"3f9b93ab-2e34-4361-8f83-873ae3112df1","resolution":{"observed_at":"2026-08-15T19:07:06.877853Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-06T20:21:23.373110Z","title":"Pryzant, et al., ”Automatic prompt optimization with large language models,” arXiv preprint arXiv:2305.03495 , 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.03223","last_updated":"2025-07-03T23:44:50Z","snapshot_observed_at":"2026-08-16T17:28:12.296798Z","submitted_at":"2025-07-03T23:44:50Z","title":"SI-Agent: An Agentic Framework for Feedback-Driven Generation and Tuning of Human-Readable System Instructions for Large Language Models","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-06T20:21:23.373110Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2507.03223"},"observation_digest":"sha256:68c37b5de01bbe847c7360b16e5e0875e12eda4ed57c67b3a81d8ba229db119a","observation_id":"c60c9bbc-260c-499f-97d2-5f75df7fd49e","resolution":{"observed_at":"2026-08-06T20:21:23.373110Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-06T18:33:05.012547Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.08109","last_updated":"2025-07-10T18:52:09Z","snapshot_observed_at":"2026-08-17T11:55:41.321805Z","submitted_at":"2025-07-10T18:52:09Z","title":"Audit, Alignment, and Optimization of LM-Powered Subroutines with Application to Public Comment Processing","version":1},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-06T18:33:05.012547Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2507.08109"},"observation_digest":"sha256:82a4e3c6d8ee4e5b6fcd72c114b27671ec9ecea27ebb1fcb229ea9a2afe4bbb5","observation_id":"447b35a4-65a8-4979-8e55-6cbd5ae0b8db","resolution":{"observed_at":"2026-08-06T18:33:05.012547Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-06T17:51:03.278384Z","title":"Automatic prompt optimization with","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.09839","last_updated":"2026-06-23T00:54:38Z","snapshot_observed_at":"2026-08-14T23:54:03.894942Z","submitted_at":"2025-07-14T00:20:14Z","title":"Stabilizing Black-Box Prompt Optimization with Textual Regularization and Signal Aggregation","version":2},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-06T17:51:03.278384Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2507.09839"},"observation_digest":"sha256:f2bb269fe6b3ce730b5b986960490b920c6b2bede81a6b7cd57c6a3fbc8c0231","observation_id":"aadb03a5-8aad-4467-8735-bfb4c5b7b56f","resolution":{"observed_at":"2026-08-06T17:51:03.278384Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-06T16:56:11.700817Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.12207","last_updated":"2025-07-16T13:07:24Z","snapshot_observed_at":"2026-08-21T08:02:07.329821Z","submitted_at":"2025-07-16T13:07:24Z","title":"BuildEvo: Designing Building Energy Consumption Forecasting Heuristics via LLM-driven Evolution","version":1},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-08-06T16:56:11.700817Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2507.12207"},"observation_digest":"sha256:ae82d7930a5cee2495359a287699d122170b3fedcf0cd71bf060977ccc5b0eea","observation_id":"720098eb-df62-4700-8594-77f97dce52e5","resolution":{"observed_at":"2026-08-06T16:56:11.700817Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-06T16:34:25.139637Z","title":"gradient descent","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2507.13158","last_updated":"2025-07-17T14:22:24Z","snapshot_observed_at":"2026-08-15T15:23:09.345799Z","submitted_at":"2025-07-17T14:22:24Z","title":"Inverse Reinforcement Learning Meets Large Language Model Post-Training: Basics, Advances, and Opportunities","version":1},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-06T16:34:25.139637Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2507.13158"},"observation_digest":"sha256:d51571b7f03a308c40d441a2146343fa82455f23bfaf2106400670fde7b32986","observation_id":"78ea5e32-c17c-4513-a199-c79c0c6fdb00","resolution":{"observed_at":"2026-08-06T16:34:25.139637Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-06T16:11:25.974242Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.15884","last_updated":"2025-07-18T21:46:15Z","snapshot_observed_at":"2026-08-16T17:28:07.666752Z","submitted_at":"2025-07-18T21:46:15Z","title":"Prompt Smart, Pay Less: Cost-Aware APO for Real-World Applications","version":1},"reference_index":2023,"source":"pdf_text","source_observed_at":"2026-08-06T16:11:25.974242Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2507.15884"},"observation_digest":"sha256:5a9c36140ae920884652af039a7bb4d2057635b4322366dae09193b28a569ded","observation_id":"840c83b5-d7dc-45b5-9c2a-6569e62b754a","resolution":{"observed_at":"2026-08-06T16:11:25.974242Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-06T14:35:46.580669Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.18618","last_updated":"2025-07-24T17:54:44Z","snapshot_observed_at":"2026-08-17T11:55:01.508052Z","submitted_at":"2025-07-24T17:54:44Z","title":"TRPrompt: Bootstrapping Query-Aware Prompt Optimization from Textual Rewards","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-06T14:35:46.580669Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2507.18618"},"observation_digest":"sha256:b46db974492b81d466cabe32f45d949cb026e88b33c48c1270edbe265c66e2ff","observation_id":"35e7ddcc-6812-40fe-ac18-ea223bc4bb2e","resolution":{"observed_at":"2026-08-06T14:35:46.580669Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-15T22:37:38.443498Z","title":"Automatic Prompt Optimization with ‘Gradient Descent’ and Beam Search,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.18638","last_updated":"2025-08-27T21:28:06Z","snapshot_observed_at":"2026-08-21T17:17:45.193066Z","submitted_at":"2025-05-10T18:27:03Z","title":"Prompt Engineering and the Effectiveness of Large Language Models in Enhancing Human Productivity","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-15T22:37:38.443498Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2507.18638"},"observation_digest":"sha256:800d0587e84b2870179abc2b8bfab06958d47b35acbe20121264a44d8be6742b","observation_id":"3cddf09b-d4f8-4c9d-85b3-669bbd6404fa","resolution":{"observed_at":"2026-08-15T22:37:38.443498Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2507.21046","last_updated":"2026-01-16T20:59:08Z","snapshot_observed_at":"2026-08-01T06:32:44.461162Z","submitted_at":"2025-07-28T17:59:05Z","title":"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence","version":4},"reference_index":43,"source":"arxiv_source","source_observed_at":"2026-05-14T22:23:14.621091Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2507.21046"},"observation_digest":"sha256:3511a67c2a10a6fd329911f83da68655dbd4694a3abed41b91f0449e26cd1752","observation_id":"2bdbb183-1cea-4b1c-b069-9490ddf2573b","resolution":{"observed_at":"2026-05-14T22:23:15.590683Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-06T00:55:38.065300Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.04086","last_updated":"2026-06-17T05:09:38Z","snapshot_observed_at":"2026-08-15T04:46:20.602692Z","submitted_at":"2025-08-06T05:04:00Z","title":"ToolGrad: Efficient Tool-use Dataset Generation with Textual \"Gradients\"","version":3},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-08-06T00:55:38.065300Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2508.04086"},"observation_digest":"sha256:9ada38e69e9881f6dfabe87ae1fef16a5cc5c7e9105dcbcaab27c177e8f38907","observation_id":"515253c4-bd57-4470-8ce0-bd2e682d4bae","resolution":{"observed_at":"2026-08-06T00:55:38.065300Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-05T23:55:53.056029Z","title":"arXiv preprint arXiv:2305.03495","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.04630","last_updated":"2025-08-06T16:57:06Z","snapshot_observed_at":"2026-08-18T01:19:53.707627Z","submitted_at":"2025-08-06T16:57:06Z","title":"CaPulse: Detecting Anomalies by Tuning in to the Causal Rhythms of Time Series","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-05T23:55:53.056029Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2508.04630"},"observation_digest":"sha256:160bf737d99c0a6802cb28e643b92249caf2cc065e6d477d68c9a04c58ac32b6","observation_id":"6e59c8d2-79bf-484e-8123-a79756763244","resolution":{"observed_at":"2026-08-05T23:55:53.056029Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-05T17:15:24.202120Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.16741","last_updated":"2025-08-22T18:33:06Z","snapshot_observed_at":"2026-08-22T06:16:22.161438Z","submitted_at":"2025-08-22T18:33:06Z","title":"WST: Weak-to-Strong Knowledge Transfer via Reinforcement Learning","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-05T17:15:24.202120Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2508.16741"},"observation_digest":"sha256:9be9c44a17b52a710dfbc1541b505964f6bd041c27992e35a964e90809ef3bd7","observation_id":"579e80bc-d222-41cb-9496-958a6861aefe","resolution":{"observed_at":"2026-08-05T17:15:24.202120Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-04T15:47:12.941106Z","title":"Automatic prompt optimization with","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2509.18384","last_updated":"2026-05-25T22:53:07Z","snapshot_observed_at":"2026-08-17T11:55:45.286321Z","submitted_at":"2025-09-22T20:14:32Z","title":"LAD-VF: LLM-Automatic Differentiation Enables Fine-Tuning-Free Robot Planning from Formal Methods Feedback","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-04T15:47:12.941106Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2509.18384"},"observation_digest":"sha256:127442868b193de1837975362a87aef71feac327c6d1ceb625257a66d87efaf7","observation_id":"2ae57cb2-65ff-4ed8-9203-d33b666837aa","resolution":{"observed_at":"2026-08-04T15:47:12.941106Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-03T20:29:15.134142Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2511.19829","last_updated":"2026-06-07T06:44:13Z","snapshot_observed_at":"2026-08-19T00:11:05.463786Z","submitted_at":"2025-11-25T01:41:13Z","title":"Knowing How to Edit: Reliable Evaluation Signals for Diagnosing and Optimizing Prompts at Query Level","version":3},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-03T20:29:15.134142Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2511.19829"},"observation_digest":"sha256:920ebdc16eaca78737fd456090ae15a14072b1c36f5a01aa920bcd20af20f5db","observation_id":"1b3e3c15-937a-493b-9528-1144b1466b80","resolution":{"observed_at":"2026-08-03T20:29:15.134142Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2512.11013","last_updated":"2026-04-07T17:30:34Z","snapshot_observed_at":"2026-08-19T21:11:44.790514Z","submitted_at":"2025-12-11T16:55:30Z","title":"PIAST: Rapid Prompting with In-context Augmentation for Scarce Training data","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-05-16T23:15:52.444217Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2512.11013"},"observation_digest":"sha256:d4e4a08a152bb1dfc2f6b456e7b303caa253ef4e7bd3a4bf19e1d9ec8b670c7c","observation_id":"a45e0833-4283-4a0e-b4ad-0791c01aac9d","resolution":{"observed_at":"2026-05-16T23:18:40.073746Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-02T23:00:27.011059Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2602.15278","last_updated":"2026-05-31T02:07:13Z","snapshot_observed_at":"2026-08-17T20:07:37.665639Z","submitted_at":"2026-02-17T00:33:53Z","title":"Visual Persuasion: What Influences Decisions of Vision-Language Models?","version":2},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-08-02T23:00:27.011059Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2602.15278"},"observation_digest":"sha256:8a1cc19944806b402c638e811f5ace851f584b498bfccf918ba8e4d4a562061c","observation_id":"51d39e98-5c40-4162-a6b7-aa2b89d2e52d","resolution":{"observed_at":"2026-08-02T23:00:27.011059Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2603.01692","last_updated":"2026-04-12T09:20:27Z","snapshot_observed_at":"2026-08-16T20:18:38.350822Z","submitted_at":"2026-03-02T10:22:47Z","title":"Reasoning as Gradient: Scaling MLE Agents Beyond Tree Search","version":3},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-05-15T17:49:46.383559Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2603.01692"},"observation_digest":"sha256:8c81950f27b9c300933a0b1e741300de3a2e6bdcb43fa835886b3be4495417ae","observation_id":"e988b4d7-58fc-499f-8bea-7ad2a6f5b85c","resolution":{"observed_at":"2026-05-15T17:50:12.234658Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-02T19:22:25.401817Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2603.02630","last_updated":"2026-05-29T15:37:29Z","snapshot_observed_at":"2026-08-02T19:22:22.341348Z","submitted_at":"2026-03-03T05:59:05Z","title":"MASPOB: Bandit-Based Prompt Optimization for Multi-Agent Systems with Graph Neural Networks","version":2},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-08-02T19:22:25.401817Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2603.02630"},"observation_digest":"sha256:b4963f9110f7026573243cb8ad9ac7bda73ee156d718b0c61cfc2ba5690ee8b3","observation_id":"34d12499-f33d-4f66-9b19-971739a86db7","resolution":{"observed_at":"2026-08-02T19:22:25.401817Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2603.06610","last_updated":"2026-05-22T08:27:37Z","snapshot_observed_at":"2026-08-19T05:05:54.748868Z","submitted_at":"2026-02-19T09:46:24Z","title":"CapTrack: Multifaceted Evaluation of Forgetting in LLM Post-Training","version":2},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-05-25T06:40:51.046965Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2603.06610"},"observation_digest":"sha256:f11527049d33ae54274a1130f37fd87be5dd22bad2f5edaaf3fad2109398a6bb","observation_id":"7429ff8e-41b7-4603-b049-3e45aec47877","resolution":{"observed_at":"2026-05-25T06:45:26.392459Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2603.28052","last_updated":"2026-03-30T05:33:50Z","snapshot_observed_at":"2026-08-11T05:18:08.192130Z","submitted_at":"2026-03-30T05:33:50Z","title":"Meta-Harness: End-to-End Optimization of Model Harnesses","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-05-13T16:15:57.877354Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2603.28052"},"observation_digest":"sha256:62c1e1dee3c82c86922083970e809b04a075143ad0b744aaf98922a9331cf779","observation_id":"fc9790a2-fa30-4e26-82b8-d49efe1a8f46","resolution":{"observed_at":"2026-05-13T16:15:58.020671Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2604.08224","last_updated":"2026-04-09T13:19:41Z","snapshot_observed_at":"2026-08-17T13:50:13.107521Z","submitted_at":"2026-04-09T13:19:41Z","title":"Externalization in LLM Agents: A Unified Review of Memory, Skills, Protocols and Harness Engineering","version":1},"reference_index":120,"source":"arxiv_source","source_observed_at":"2026-05-10T17:40:14.733882Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2604.08224"},"observation_digest":"sha256:b2cdc1033d6366b5824db204b04b5bf7441ed2dece7793705639c1795df03f54","observation_id":"7054d502-1b0e-4239-9319-1c9aacb3f196","resolution":{"observed_at":"2026-05-11T06:20:59.723082Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2604.14585","last_updated":"2026-05-27T06:20:29Z","snapshot_observed_at":"2026-08-15T11:50:23.792579Z","submitted_at":"2026-04-16T03:23:46Z","title":"Prompt Optimization Is a Coin Flip: Diagnosing When It Helps in Compound AI Systems","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-05-10T11:18:09.127345Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2604.14585"},"observation_digest":"sha256:e962ae88cf62cef96545bb3cc7589f3b5675cc7e5783b0ce9085d7806dab1638","observation_id":"654bae84-0984-423c-84b7-af515b9411f0","resolution":{"observed_at":"2026-05-10T11:20:10.383973Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2605.15803","last_updated":"2026-05-15T09:56:40Z","snapshot_observed_at":"2026-08-16T20:50:43.622330Z","submitted_at":"2026-05-15T09:56:40Z","title":"Embedding-perturbed Exploration Preference Optimization for Flow Models","version":1},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-05-20T18:33:52.933672Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2605.15803"},"observation_digest":"sha256:7e3fdd8b0aeb5a3f79e715765d71576d415b51cd62fc8d05cb79349fae2d4fae","observation_id":"90dd37db-fc11-464d-91bb-08891c753f77","resolution":{"observed_at":"2026-05-20T18:38:53.027822Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2605.18780","last_updated":"2026-04-29T06:13:39Z","snapshot_observed_at":"2026-08-13T02:06:14.738653Z","submitted_at":"2026-04-29T06:13:39Z","title":"A Reproducibility Analysis of PO4ISR: Diagnosing and Mitigating Semantic Drift in LLM-Based Session Recommendation","version":1},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-05-21T00:45:14.845401Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2605.18780"},"observation_digest":"sha256:98bebf5e97872151fa2f25a68c6d6248a25d63a7d540c68a335769d2669719e5","observation_id":"4c9b3fa0-77a0-4645-8abb-f281dd2f23ce","resolution":{"observed_at":"2026-05-21T00:49:19.533621Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2606.08867","last_updated":"2026-06-07T22:44:00Z","snapshot_observed_at":"2026-08-03T19:07:50.944087Z","submitted_at":"2026-06-07T22:44:00Z","title":"Building Customer Support AI Agents at 100M-User Scale: An Evaluation-Driven Framework","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-06-27T18:21:57.096578Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2606.08867"},"observation_digest":"sha256:162361ab8efa58b4790b702026532e1ea0119b9b3211984806ba07a221814ad7","observation_id":"f7bd9f60-3431-46c1-8a9c-a26c023ae623","resolution":{"observed_at":"2026-07-02T23:17:29.499369Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2606.12387","last_updated":"2026-06-10T17:52:15Z","snapshot_observed_at":"2026-08-09T09:51:44.818603Z","submitted_at":"2026-06-10T17:52:15Z","title":"TAHOE: Text-to-SQL with Automated Hint Optimization from Experience","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-06-27T07:28:32.185178Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2606.12387"},"observation_digest":"sha256:137100ecceacf49d8f280073359a28c5e5ad2aa9c911276b2133f629d6936a5b","observation_id":"a39b3b0b-5d08-4fb1-84a1-411370f9e410","resolution":{"observed_at":"2026-07-03T13:48:21.407189Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2606.18620","last_updated":"2026-06-17T02:34:32Z","snapshot_observed_at":"2026-08-06T17:11:29.341597Z","submitted_at":"2026-06-17T02:34:32Z","title":"BCL: Bayesian In-Context Learning Framework for Information Extraction","version":1},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-06-26T21:11:23.042546Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2606.18620"},"observation_digest":"sha256:6585ca31729b7d59b747279335080c463e15a80e2163f110f1ecb2249cf01b0d","observation_id":"f4f18d23-9a3e-489c-bb8d-7cc519028ba9","resolution":{"observed_at":"2026-07-04T00:29:15.908307Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2606.20475","last_updated":"2026-06-18T16:54:25Z","snapshot_observed_at":"2026-08-17T00:46:48.934209Z","submitted_at":"2026-06-18T16:54:25Z","title":"Marginal Advantage Accumulation for Memory-Driven Agent Self-Evolution","version":1},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-06-26T17:45:49.272070Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2606.20475"},"observation_digest":"sha256:ef7cc2aac0844464b4d555ba248869f0c9c49d04e1b88a898a9300b571aab677","observation_id":"adb0eadf-fa52-4a37-a0e6-c8ed9cb663c6","resolution":{"observed_at":"2026-07-04T03:39:30.821536Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2606.22385","last_updated":"2026-06-21T08:22:23Z","snapshot_observed_at":"2026-08-01T18:23:09.613202Z","submitted_at":"2026-06-21T08:22:23Z","title":"MetaPS: Adaptive Programmatic Strategy Selection for Market Agents","version":1},"reference_index":68,"source":"arxiv_source","source_observed_at":"2026-06-26T11:06:28.690956Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2606.22385"},"observation_digest":"sha256:ae5aaa36dc6f73635bc98f2b1461b7ed5130feae4b3bb6fc42216c863a7f186b","observation_id":"061943ce-084a-435a-bdf3-68774a070a4f","resolution":{"observed_at":"2026-07-04T08:39:42.748336Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2606.22803","last_updated":"2026-06-23T02:56:13Z","snapshot_observed_at":"2026-08-17T11:38:58.149323Z","submitted_at":"2026-06-22T03:26:08Z","title":"Towards Fast Domain Adaptation and Fine-Grained User Simulation for Evaluating Conversational Recommender Systems","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-06-26T07:13:21.198407Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2606.22803"},"observation_digest":"sha256:e7ef56c725d7672aa7bc33640c04a02323cf7be060df20644ef1298b5449df93","observation_id":"4fdfb860-2707-41d9-90b3-dcd53d517e61","resolution":{"observed_at":"2026-07-04T12:09:48.813322Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2606.25193","last_updated":"2026-06-23T21:36:42Z","snapshot_observed_at":"2026-08-17T04:20:57.558038Z","submitted_at":"2026-06-23T21:36:42Z","title":"LLM4MTLs: Automated Generation and Empirical Evaluation of Model Transformation Languages","version":1},"reference_index":50,"source":"arxiv_source","source_observed_at":"2026-06-25T22:38:37.806421Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2606.25193"},"observation_digest":"sha256:da1c076e2a1a647cf0d341f9111e73d27f6ad9dd5070236185f7fbede7bb45b0","observation_id":"63fd08e3-f085-46b9-920e-49040285988e","resolution":{"observed_at":"2026-07-04T18:40:03.104242Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":"2305.03495","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-04T18:40:03.102880Z","title":"Rafailov, R., Sharma, A., Mitchell, E., Ermon, S., Manning, C","venue":null,"work_id":"8f5910e5-bea1-4761-87ec-05f692dd6f04","year":2023},"citing_paper":{"arxiv_id":"2607.00427","last_updated":"2026-07-01T04:36:30Z","snapshot_observed_at":"2026-08-14T21:26:49.756857Z","submitted_at":"2026-07-01T04:36:30Z","title":"BT-APE: A Computationally Light Backtracking Approach to Automatic Prompt Engineering for Requirements Classification","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-07-02T09:12:59.039529Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2607.00427"},"observation_digest":"sha256:b7b01d5e32bd8a19ddcb5e643cac7aa70d96827059849e68b2a89d214d90b012","observation_id":"d26e4894-fb73-4ea1-afc8-99341156c044","resolution":{"observed_at":"2026-07-02T09:16:48.666443Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-22T06:32:14.747728+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-07-14T04:44:42.371749Z","title":"gradient descent","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.11565","last_updated":"2026-07-13T13:47:17Z","snapshot_observed_at":"2026-08-15T11:55:24.074429Z","submitted_at":"2026-07-13T13:47:17Z","title":"Heuristic Learning for Active Flow Control Using Coding Agents","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-07-14T04:44:42.371749Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2607.11565"},"observation_digest":"sha256:d7b70c1da9c8e3f867e99ddb016100a94f4e32a6986ca6800c0ad1dbdb1654ad","observation_id":"faa636a4-b382-437e-bdca-723a4f9ea488","resolution":{"observed_at":"2026-07-14T04:44:42.371749Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-02T05:43:00.443189Z","title":"Automatic prompt optimization with ``gradient descent'' and beam search, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.14159","last_updated":"2026-07-14T21:22:18Z","snapshot_observed_at":"2026-08-18T12:32:44.084161Z","submitted_at":"2026-07-14T21:22:18Z","title":"MemoHarness: Agent Harnesses That Learn from Experience","version":1},"reference_index":54,"source":"arxiv_source","source_observed_at":"2026-08-02T05:43:00.443189Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2607.14159"},"observation_digest":"sha256:0982e121ff302eb1da106c1e56639417c7bcd09783b0c43845abf795a03401dd","observation_id":"5a966f4a-b337-4348-b2d5-2704c541d164","resolution":{"observed_at":"2026-08-02T05:43:00.443189Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-06T00:20:37.207547Z","title":"Gradient Descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2608.01366","last_updated":"2026-08-04T05:28:54Z","snapshot_observed_at":"2026-08-15T01:10:55.224631Z","submitted_at":"2026-08-02T16:33:27Z","title":"Asking Questions the Right Way: A Multi-Agent Conversational System for Prompt Formulation in Complex Task Resolution","version":2},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-06T00:20:37.207547Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2608.01366"},"observation_digest":"sha256:b1775662f1da8cb807db5fb23f3839d2c4e76dc68de300ca25bceaa35d2aa758","observation_id":"3cffb090-0a89-462c-be0c-f4ea62d39747","resolution":{"observed_at":"2026-08-06T00:20:37.207547Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-04T10:10:32.578208Z","title":"Qwen Team","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.02276","last_updated":"2026-08-03T14:12:18Z","snapshot_observed_at":"2026-08-18T15:29:04.959048Z","submitted_at":"2026-08-03T14:12:18Z","title":"Harness-R1: Learning to Edit Executable Runtime Harnesses from Agent Failure Trajectories","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-04T10:10:32.578208Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2608.02276"},"observation_digest":"sha256:f37e3ea77ba2be4891d79581146d5f2f87c197758331dc509f152fb7c2fe035d","observation_id":"cb3fabb6-421b-4749-a9a1-4ba47e3aa620","resolution":{"observed_at":"2026-08-04T10:10:32.578208Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-15T14:58:05.734796Z","title":"Gradient Descent","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.02919","last_updated":"2026-08-03T22:13:12Z","snapshot_observed_at":"2026-08-20T08:08:26.305019Z","submitted_at":"2026-08-03T22:13:12Z","title":"FLARE: Few-shot Learning-based Adaptive Reflective Engine","version":1},"reference_index":2023,"source":"pdf_text","source_observed_at":"2026-08-15T14:58:05.734796Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2608.02919"},"observation_digest":"sha256:409eb2978f1301a8431d824903518946c11352b8a1d86abfb21303e6a655b423","observation_id":"fe5e747e-cfbe-4cb1-86b1-f8cb6e2dcb3f","resolution":{"observed_at":"2026-08-15T14:58:05.734796Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-15T14:00:07.898848Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2608.11513","last_updated":"2026-08-11T23:57:06Z","snapshot_observed_at":"2026-08-18T10:39:33.774546Z","submitted_at":"2026-08-11T23:57:06Z","title":"Do Influence Tactics Matter? Investigating Prompt Framing Effects in LLM Code Generation","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-15T14:00:07.898848Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2608.11513"},"observation_digest":"sha256:b4257dd98a786863c09fc3470f2c2271d86c9b08233f89b0ade4e292d16d0288","observation_id":"dc143e12-766b-43b8-8e60-a00c8a8b140c","resolution":{"observed_at":"2026-08-15T14:00:07.898848Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03495","snapshot_observed_at":"2026-08-16T00:39:52.369060Z","title":"gradient descent","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2608.11694","last_updated":"2026-08-12T06:05:47Z","snapshot_observed_at":"2026-08-19T09:17:49.448595Z","submitted_at":"2026-08-12T06:05:47Z","title":"The Wording Effect: Quantifying Two-Way Drift in LLM Benchmark Performance","version":1},"reference_index":95,"source":"arxiv_source","source_observed_at":"2026-08-16T00:39:52.369060Z"},"links":{"cited_paper":"/paper/2305.03495","citing_paper":"/paper/2608.11694"},"observation_digest":"sha256:fb50fc56ac6e798672b0d81176e611745bd4c938c355e11a97d39106adc4c5dd","observation_id":"71c17b2b-a50e-4d4a-9cdd-a44fb3bd3a75","resolution":{"observed_at":"2026-08-16T00:39:52.369060Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2305.03495/citation-record","integrity":"/paper/2305.03495/integrity","json":"/paper/2305.03495/citation-record.json","paper":"/paper/2305.03495"},"outbound":[],"paper":{"arxiv_id":"2305.03495","last_updated":"2023-10-19T04:37:25Z","latest_version":2,"primary_category":"cs.CL","snapshot_observed_at":"2026-08-16T15:35:36.835145Z","submitted_at":"2023-05-04T15:15:22Z","title":"Automatic Prompt Optimization with \"Gradient Descent\" and Beam Search"},"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":"2026-08-22T06:32:14.747728+00:00","source":"crossref"},{"observed_at":"2026-08-22T06:32:06.552537+00:00","source":"retraction_watch"}],"thesis":"As of 22 August 2026, this Paper Citation Record lists 0 of 0 outbound references and 69 inbound Pith citation observations for arXiv:2305.03495."}