{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KM6I5WOYI62IV2PGCSXBKUOA4O","short_pith_number":"pith:KM6I5WOY","schema_version":"1.0","canonical_sha256":"533c8ed9d847b48ae9e614ae1551c0e3be904b9194d39bfe803102a5ff2eb1e1","source":{"kind":"arxiv","id":"2603.16856","version":2},"attestation_state":"computed","paper":{"title":"Online Experiential Learning for Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Furu Wei, Li Dong, Qingxiu Dong, Shaohan Huang, Tianzhu Ye, Xun Wu","submitted_at":"2026-03-17T17:57:49Z","abstract_excerpt":"The prevailing paradigm for improving large language models relies on offline training with human annotations or simulated environments, leaving the rich experience accumulated during real-world deployment entirely unexploited. We propose Online Experiential Learning (OEL), a framework that enables language models to continuously improve from their own deployment experience. OEL operates in two stages: first, transferable experiential knowledge is extracted and accumulated from interaction trajectories collected on the user side; second, this knowledge is consolidated into model parameters via"},"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":"2603.16856","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-03-17T17:57:49Z","cross_cats_sorted":[],"title_canon_sha256":"66e42fddafb785c47423bd4c159583ce31c313c523482ba43773e74d1e17a381","abstract_canon_sha256":"8f4cef8aa8834b59d58b873498494aa150e62127aae3a1b20427312bf4907eba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:17:18.441809Z","signature_b64":"Oaj5K9YmXYJlBATRpjr7vGELgCVP4nAlqh9uGDVHnqjhMKi55fxaioQhLjLr2xLYal1sr1E8js9qnMC6FGPeBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"533c8ed9d847b48ae9e614ae1551c0e3be904b9194d39bfe803102a5ff2eb1e1","last_reissued_at":"2026-06-30T02:17:18.441171Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:17:18.441171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Online Experiential Learning for Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Furu Wei, Li Dong, Qingxiu Dong, Shaohan Huang, Tianzhu Ye, Xun Wu","submitted_at":"2026-03-17T17:57:49Z","abstract_excerpt":"The prevailing paradigm for improving large language models relies on offline training with human annotations or simulated environments, leaving the rich experience accumulated during real-world deployment entirely unexploited. We propose Online Experiential Learning (OEL), a framework that enables language models to continuously improve from their own deployment experience. OEL operates in two stages: first, transferable experiential knowledge is extracted and accumulated from interaction trajectories collected on the user side; second, this knowledge is consolidated into model parameters via"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.16856","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/2603.16856/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":"2603.16856","created_at":"2026-06-30T02:17:18.441245+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.16856v2","created_at":"2026-06-30T02:17:18.441245+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.16856","created_at":"2026-06-30T02:17:18.441245+00:00"},{"alias_kind":"pith_short_12","alias_value":"KM6I5WOYI62I","created_at":"2026-06-30T02:17:18.441245+00:00"},{"alias_kind":"pith_short_16","alias_value":"KM6I5WOYI62IV2PG","created_at":"2026-06-30T02:17:18.441245+00:00"},{"alias_kind":"pith_short_8","alias_value":"KM6I5WOY","created_at":"2026-06-30T02:17:18.441245+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":13,"sample":[{"citing_arxiv_id":"2605.30070","citing_title":"A Predictive Law for On-Policy Self-Distillation From World Feedback","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18141","citing_title":"A Brief Overview: On-Policy Self-Distillation In Large Language Models","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22263","citing_title":"Tailoring Teaching to Aptitude: Direction-Adaptive Self-Distillation for LLM Reasoning","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21984","citing_title":"Echo: Learning from Experience Data via User-Driven Refinement","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2605.16363","citing_title":"ORACLE: Anticipating Scams from Partial Trajectories in Streaming App Usage","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2605.15113","citing_title":"Learning from Language Feedback via Variational Policy Distillation","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18141","citing_title":"A Brief Overview: On-Policy Self-Distillation In Large Language Models","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2605.19444","citing_title":"Detecting and Mitigating the Correct-Answer Extinction Window in Test-Time Reinforcement Learning with Majority Voting","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06130","citing_title":"Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09192","citing_title":"Evidence Over Plans: Online Trajectory Verification for Skill Distillation","ref_index":23,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06130","citing_title":"Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10674","citing_title":"Skill-SD: Skill-Conditioned Self-Distillation for Multi-turn LLM Agents","ref_index":40,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06130","citing_title":"Skill1: Unified Evolution of Skill-Augmented Agents via Reinforcement Learning","ref_index":24,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KM6I5WOYI62IV2PGCSXBKUOA4O","json":"https://pith.science/pith/KM6I5WOYI62IV2PGCSXBKUOA4O.json","graph_json":"https://pith.science/api/pith-number/KM6I5WOYI62IV2PGCSXBKUOA4O/graph.json","events_json":"https://pith.science/api/pith-number/KM6I5WOYI62IV2PGCSXBKUOA4O/events.json","paper":"https://pith.science/paper/KM6I5WOY"},"agent_actions":{"view_html":"https://pith.science/pith/KM6I5WOYI62IV2PGCSXBKUOA4O","download_json":"https://pith.science/pith/KM6I5WOYI62IV2PGCSXBKUOA4O.json","view_paper":"https://pith.science/paper/KM6I5WOY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.16856&json=true","fetch_graph":"https://pith.science/api/pith-number/KM6I5WOYI62IV2PGCSXBKUOA4O/graph.json","fetch_events":"https://pith.science/api/pith-number/KM6I5WOYI62IV2PGCSXBKUOA4O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KM6I5WOYI62IV2PGCSXBKUOA4O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KM6I5WOYI62IV2PGCSXBKUOA4O/action/storage_attestation","attest_author":"https://pith.science/pith/KM6I5WOYI62IV2PGCSXBKUOA4O/action/author_attestation","sign_citation":"https://pith.science/pith/KM6I5WOYI62IV2PGCSXBKUOA4O/action/citation_signature","submit_replication":"https://pith.science/pith/KM6I5WOYI62IV2PGCSXBKUOA4O/action/replication_record"}},"created_at":"2026-06-30T02:17:18.441245+00:00","updated_at":"2026-06-30T02:17:18.441245+00:00"}