{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HYLCUSLTMCGMJIHVETPZ4PT4HA","short_pith_number":"pith:HYLCUSLT","schema_version":"1.0","canonical_sha256":"3e162a4973608cc4a0f524df9e3e7c383faaa93962c7030eb080f7d6b00def51","source":{"kind":"arxiv","id":"2410.23214","version":2},"attestation_state":"computed","paper":{"title":"Grounding by Trying: LLMs with Reinforcement Learning-Enhanced Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Archit Sharma, Chelsea Finn, Omar Khattab, Sheryl Hsu","submitted_at":"2024-10-30T17:02:54Z","abstract_excerpt":"The hallucinations of large language models (LLMs) are increasingly mitigated by allowing LLMs to search for information and to ground their answers in real sources. Unfortunately, LLMs often struggle with posing the right search queries, especially when dealing with complex or otherwise indirect topics. Observing that LLMs can learn to search for relevant facts by $\\textit{trying}$ different queries and learning to up-weight queries that successfully produce relevant results, we introduce $\\underline{Le}$arning to $\\underline{Re}$trieve by $\\underline{T}$rying (LeReT), a reinforcement learnin"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2410.23214","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-30T17:02:54Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"52a9b7db5cb23986215e0823c1edd01b68715f618e7ee70bd140cd6d8bbc3a90","abstract_canon_sha256":"593ffb9f5d426e0a4dd2b07e7e2214b3cf6a96a85baa56d06d4a8eada669dac7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:28:57.740311Z","signature_b64":"7awOmAp5Sgn5XIzxtpRI1wah6gUUJNpBSyIx5ApF2eZi1z5N34lpJbSKz+MGMZimqcdIsLPoaiXdp+PkGTKgCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3e162a4973608cc4a0f524df9e3e7c383faaa93962c7030eb080f7d6b00def51","last_reissued_at":"2026-07-05T09:28:57.739775Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:28:57.739775Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Grounding by Trying: LLMs with Reinforcement Learning-Enhanced Retrieval","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Archit Sharma, Chelsea Finn, Omar Khattab, Sheryl Hsu","submitted_at":"2024-10-30T17:02:54Z","abstract_excerpt":"The hallucinations of large language models (LLMs) are increasingly mitigated by allowing LLMs to search for information and to ground their answers in real sources. Unfortunately, LLMs often struggle with posing the right search queries, especially when dealing with complex or otherwise indirect topics. Observing that LLMs can learn to search for relevant facts by $\\textit{trying}$ different queries and learning to up-weight queries that successfully produce relevant results, we introduce $\\underline{Le}$arning to $\\underline{Re}$trieve by $\\underline{T}$rying (LeReT), a reinforcement learnin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.23214","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2410.23214/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2410.23214","created_at":"2026-07-05T09:28:57.739832+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.23214v2","created_at":"2026-07-05T09:28:57.739832+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.23214","created_at":"2026-07-05T09:28:57.739832+00:00"},{"alias_kind":"pith_short_12","alias_value":"HYLCUSLTMCGM","created_at":"2026-07-05T09:28:57.739832+00:00"},{"alias_kind":"pith_short_16","alias_value":"HYLCUSLTMCGMJIHV","created_at":"2026-07-05T09:28:57.739832+00:00"},{"alias_kind":"pith_short_8","alias_value":"HYLCUSLT","created_at":"2026-07-05T09:28:57.739832+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.01248","citing_title":"$S^3$-R1: Learning to Retrieve and Answer Step-by-Step with Synthetic Data","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2502.13957","citing_title":"Supervising the search process produces reliable and generalizable information-seeking agents","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2508.15411","citing_title":"Foundational Design Principles and Patterns for Building Robust and Adaptive GenAI-Native Systems","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01248","citing_title":"$S^3$-R1: Learning to Retrieve and Answer Step-by-Step with Synthetic Data","ref_index":9,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HYLCUSLTMCGMJIHVETPZ4PT4HA","json":"https://pith.science/pith/HYLCUSLTMCGMJIHVETPZ4PT4HA.json","graph_json":"https://pith.science/api/pith-number/HYLCUSLTMCGMJIHVETPZ4PT4HA/graph.json","events_json":"https://pith.science/api/pith-number/HYLCUSLTMCGMJIHVETPZ4PT4HA/events.json","paper":"https://pith.science/paper/HYLCUSLT"},"agent_actions":{"view_html":"https://pith.science/pith/HYLCUSLTMCGMJIHVETPZ4PT4HA","download_json":"https://pith.science/pith/HYLCUSLTMCGMJIHVETPZ4PT4HA.json","view_paper":"https://pith.science/paper/HYLCUSLT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.23214&json=true","fetch_graph":"https://pith.science/api/pith-number/HYLCUSLTMCGMJIHVETPZ4PT4HA/graph.json","fetch_events":"https://pith.science/api/pith-number/HYLCUSLTMCGMJIHVETPZ4PT4HA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HYLCUSLTMCGMJIHVETPZ4PT4HA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HYLCUSLTMCGMJIHVETPZ4PT4HA/action/storage_attestation","attest_author":"https://pith.science/pith/HYLCUSLTMCGMJIHVETPZ4PT4HA/action/author_attestation","sign_citation":"https://pith.science/pith/HYLCUSLTMCGMJIHVETPZ4PT4HA/action/citation_signature","submit_replication":"https://pith.science/pith/HYLCUSLTMCGMJIHVETPZ4PT4HA/action/replication_record"}},"created_at":"2026-07-05T09:28:57.739832+00:00","updated_at":"2026-07-05T09:28:57.739832+00:00"}