{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:EBG36OE2G4KEDL4YFPCKOLPLTL","short_pith_number":"pith:EBG36OE2","schema_version":"1.0","canonical_sha256":"204dbf389a371441af982bc4a72deb9ace432495a8bc78039dcd7ad1e0e7b82c","source":{"kind":"arxiv","id":"2607.26094","version":1},"attestation_state":"computed","paper":{"title":"Meta-Learned Reward Shaping for Reinforcement Learning from Human Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Yunpeng Chu","submitted_at":"2026-07-28T02:09:26Z","abstract_excerpt":"Reinforcement Learning from Human Feedback (RLHF) is the standard approach for aligning large language models with human preferences, but its quality is limited by static, task-agnostic reward models. This mismatch leads to sparse learning signals and suboptimal alignment. We introduce MeRLa (Meta-Learned Reward Shaping), a principled framework that meta-learns a task-aware shaping function $\\Phi(x,y;\\phi)$ across auxiliary tasks before RLHF training. The learned shaping produces a composite reward that preserves policy optimality while providing task-specific learning signals. Our meta-object"},"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":"2607.26094","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-07-28T02:09:26Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"f9c23e54e1019ef53cb95702c9fe0d9c766a0ee5a34db211b1cbf7c986c4fac2","abstract_canon_sha256":"3f84db8fca84d4261cc9ae497fd82ebf6b04e1ed6c98f18e0a927c14c000af03"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"204dbf389a371441af982bc4a72deb9ace432495a8bc78039dcd7ad1e0e7b82c","last_reissued_at":"2026-07-30T00:08:20.269544Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T00:08:20.269544Z"},"graph_snapshot":{"paper":{"title":"Meta-Learned Reward Shaping for Reinforcement Learning from Human Feedback","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Yunpeng Chu","submitted_at":"2026-07-28T02:09:26Z","abstract_excerpt":"Reinforcement Learning from Human Feedback (RLHF) is the standard approach for aligning large language models with human preferences, but its quality is limited by static, task-agnostic reward models. This mismatch leads to sparse learning signals and suboptimal alignment. We introduce MeRLa (Meta-Learned Reward Shaping), a principled framework that meta-learns a task-aware shaping function $\\Phi(x,y;\\phi)$ across auxiliary tasks before RLHF training. The learned shaping produces a composite reward that preserves policy optimality while providing task-specific learning signals. Our meta-object"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26094","kind":"arxiv","version":1},"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/2607.26094/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":"2607.26094","created_at":"2026-07-30T00:08:20.273152+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.26094v1","created_at":"2026-07-30T00:08:20.273152+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26094","created_at":"2026-07-30T00:08:20.273152+00:00"},{"alias_kind":"pith_short_12","alias_value":"EBG36OE2G4KE","created_at":"2026-07-30T00:08:20.273152+00:00"},{"alias_kind":"pith_short_16","alias_value":"EBG36OE2G4KEDL4Y","created_at":"2026-07-30T00:08:20.273152+00:00"},{"alias_kind":"pith_short_8","alias_value":"EBG36OE2","created_at":"2026-07-30T00:08:20.273152+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EBG36OE2G4KEDL4YFPCKOLPLTL","json":"https://pith.science/pith/EBG36OE2G4KEDL4YFPCKOLPLTL.json","graph_json":"https://pith.science/api/pith-number/EBG36OE2G4KEDL4YFPCKOLPLTL/graph.json","events_json":"https://pith.science/api/pith-number/EBG36OE2G4KEDL4YFPCKOLPLTL/events.json","paper":"https://pith.science/paper/EBG36OE2"},"agent_actions":{"view_html":"https://pith.science/pith/EBG36OE2G4KEDL4YFPCKOLPLTL","download_json":"https://pith.science/pith/EBG36OE2G4KEDL4YFPCKOLPLTL.json","view_paper":"https://pith.science/paper/EBG36OE2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.26094&json=true","fetch_graph":"https://pith.science/api/pith-number/EBG36OE2G4KEDL4YFPCKOLPLTL/graph.json","fetch_events":"https://pith.science/api/pith-number/EBG36OE2G4KEDL4YFPCKOLPLTL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EBG36OE2G4KEDL4YFPCKOLPLTL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EBG36OE2G4KEDL4YFPCKOLPLTL/action/storage_attestation","attest_author":"https://pith.science/pith/EBG36OE2G4KEDL4YFPCKOLPLTL/action/author_attestation","sign_citation":"https://pith.science/pith/EBG36OE2G4KEDL4YFPCKOLPLTL/action/citation_signature","submit_replication":"https://pith.science/pith/EBG36OE2G4KEDL4YFPCKOLPLTL/action/replication_record"}},"created_at":"2026-07-30T00:08:20.273152+00:00","updated_at":"2026-07-30T00:08:20.273152+00:00"}