{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5PU2B5CYD36UGQXFRFF2AK44SX","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":"d9de390264ae001973350a5bc7dfc249a19539e73d028703049d746619ea4ce1","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T17:58:02Z","title_canon_sha256":"d33ff6f88be2fb81a07880dc6dfbb8f6f4d758ccf40e503078120713955c19fb"},"schema_version":"1.0","source":{"id":"2502.18407","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.18407","created_at":"2026-07-05T10:19:53Z"},{"alias_kind":"arxiv_version","alias_value":"2502.18407v1","created_at":"2026-07-05T10:19:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.18407","created_at":"2026-07-05T10:19:53Z"},{"alias_kind":"pith_short_12","alias_value":"5PU2B5CYD36U","created_at":"2026-07-05T10:19:53Z"},{"alias_kind":"pith_short_16","alias_value":"5PU2B5CYD36UGQXF","created_at":"2026-07-05T10:19:53Z"},{"alias_kind":"pith_short_8","alias_value":"5PU2B5CY","created_at":"2026-07-05T10:19:53Z"}],"graph_snapshots":[{"event_id":"sha256:c1df9c2472312b415564beaba44c97a5162991aa360316e0bd0ca50734c621e1","target":"graph","created_at":"2026-07-05T10:19:53Z","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/2502.18407/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Existing LLM-based agents have achieved strong performance on held-in tasks, but their generalizability to unseen tasks remains poor. Hence, some recent work focus on fine-tuning the policy model with more diverse tasks to improve the generalizability. In this work, we find that finetuning a reward model to guide the policy model is more robust than directly finetuning the policy model. Based on this finding, we propose AgentRM, a generalizable reward model, to guide the policy model for effective test-time search. We comprehensively investigate three approaches to construct the reward model, ","authors_text":"Jingru Fan, Maosong Sun, Siyu Yan, Weize Chen, Xin Cong, Yankai Lin, Yaxi Lu, Yu Xia, Zhiyuan Liu, Zhong Zhang","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T17:58:02Z","title":"AgentRM: Enhancing Agent Generalization with Reward Modeling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.18407","kind":"arxiv","version":1},"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":"sha256:000d9fe67727d01a1199c4b73de74323d3d492e3725fc0322e4577c586b271dd","target":"record","created_at":"2026-07-05T10:19:53Z","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":"d9de390264ae001973350a5bc7dfc249a19539e73d028703049d746619ea4ce1","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T17:58:02Z","title_canon_sha256":"d33ff6f88be2fb81a07880dc6dfbb8f6f4d758ccf40e503078120713955c19fb"},"schema_version":"1.0","source":{"id":"2502.18407","kind":"arxiv","version":1}},"canonical_sha256":"ebe9a0f4581efd4342e5894ba02b9c95e76f59c96d3a5b4dd24bcfe05808b8ec","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ebe9a0f4581efd4342e5894ba02b9c95e76f59c96d3a5b4dd24bcfe05808b8ec","first_computed_at":"2026-07-05T10:19:53.479660Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:19:53.479660Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"87cgRdNjIkdu7zAv/a3kERQV/apzkZc5oHJ7su6twv8+6JFZxal9FN2NyHeikdhxt+g4bjzwwl8mG3Ux/joCBA==","signature_status":"signed_v1","signed_at":"2026-07-05T10:19:53.480127Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.18407","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:000d9fe67727d01a1199c4b73de74323d3d492e3725fc0322e4577c586b271dd","sha256:c1df9c2472312b415564beaba44c97a5162991aa360316e0bd0ca50734c621e1"],"state_sha256":"dd1763cac89665d57d1bd94daa879f12cb341a926b32a79fd53c3f7333a863e1"}