{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:ANTNBXEPD5XYWVARWCPT7J3JP4","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"a4c168918de022ab22d2a96d970cf8b7ac18ea5412aa5c47186fea0b03a48b8f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-28T21:18:49Z","title_canon_sha256":"52b722f9c6cf906a8d8b96c47329d07e86e8515241b0cbc281b5300d107b402e"},"schema_version":"1.0","source":{"id":"2410.21545","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.21545","created_at":"2026-07-05T10:15:46Z"},{"alias_kind":"arxiv_version","alias_value":"2410.21545v2","created_at":"2026-07-05T10:15:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.21545","created_at":"2026-07-05T10:15:46Z"},{"alias_kind":"pith_short_12","alias_value":"ANTNBXEPD5XY","created_at":"2026-07-05T10:15:46Z"},{"alias_kind":"pith_short_16","alias_value":"ANTNBXEPD5XYWVAR","created_at":"2026-07-05T10:15:46Z"},{"alias_kind":"pith_short_8","alias_value":"ANTNBXEP","created_at":"2026-07-05T10:15:46Z"}],"graph_snapshots":[{"event_id":"sha256:6b0ad4339c944e2cbb6b912b038f74c60b4d30a7b36db2450a2d5a434fb269c3","target":"graph","created_at":"2026-07-05T10:15:46Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2410.21545/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reward modeling in large language models is susceptible to reward hacking, causing models to latch onto superficial features such as the tendency to generate lists or unnecessarily long responses. In reinforcement learning from human feedback (RLHF) and more generally during post-training flawed reward signals often lead to outputs that optimize for these spurious correlates instead of genuine quality or correctness. We propose Context-Aware Reward Modeling (CARMO), a novel approach that first generates dynamic, context-relevant criteria to ground the reward model before producing reward score","authors_text":"Chetan Bansal, Huaxiu Yao, Rahul Madhavan, Saravan Rajmohan, Shivam Shandilya, Supriyo Ghosh, Taneesh Gupta, Xuchao Zhang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-28T21:18:49Z","title":"CARMO: Dynamic Criteria Generation for Context-Aware Reward Modelling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.21545","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8a333e576e9b48899137b7495257a4f6cbd2f3d26332c351063b9fb7bc514f5e","target":"record","created_at":"2026-07-05T10:15:46Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"a4c168918de022ab22d2a96d970cf8b7ac18ea5412aa5c47186fea0b03a48b8f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-10-28T21:18:49Z","title_canon_sha256":"52b722f9c6cf906a8d8b96c47329d07e86e8515241b0cbc281b5300d107b402e"},"schema_version":"1.0","source":{"id":"2410.21545","kind":"arxiv","version":2}},"canonical_sha256":"0366d0dc8f1f6f8b5411b09f3fa7697f05c69adca67a99e35cb6fc4634c8f45a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0366d0dc8f1f6f8b5411b09f3fa7697f05c69adca67a99e35cb6fc4634c8f45a","first_computed_at":"2026-07-05T10:15:46.982836Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:15:46.982836Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+DocquoiqegXIlLBFFh033wwjxm0YlB0wa2H25+HwjhHdFl/hlMMaOLdYwxFyil5NmRuakhH1IJkGGkD9R4GCw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:15:46.983447Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.21545","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8a333e576e9b48899137b7495257a4f6cbd2f3d26332c351063b9fb7bc514f5e","sha256:6b0ad4339c944e2cbb6b912b038f74c60b4d30a7b36db2450a2d5a434fb269c3"],"state_sha256":"fcf06ebe94b9f894044a09c69f51a95bc07d727e3287e44ce23b1416a1cd6afb"}