{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:ED2RYF65QGFTRD66ZGFEX24RZJ","short_pith_number":"pith:ED2RYF65","canonical_record":{"source":{"id":"2310.01045","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-02T09:47:40Z","cross_cats_sorted":[],"title_canon_sha256":"0d4fb0b0c1eca6497debcd8f079e31467e39a5a66410ceb95340b0797facdb1d","abstract_canon_sha256":"490bc3f32088d4624d7c113720c9add78b365af3ab6fa4c2407af354d2ea9b17"},"schema_version":"1.0"},"canonical_sha256":"20f51c17dd818b388fdec98a4beb91ca5108f364cce15a0175022ab02dc2b0a5","source":{"kind":"arxiv","id":"2310.01045","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.01045","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"arxiv_version","alias_value":"2310.01045v2","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.01045","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"pith_short_12","alias_value":"ED2RYF65QGFT","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"pith_short_16","alias_value":"ED2RYF65QGFTRD66","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"pith_short_8","alias_value":"ED2RYF65","created_at":"2026-07-05T07:43:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:ED2RYF65QGFTRD66ZGFEX24RZJ","target":"record","payload":{"canonical_record":{"source":{"id":"2310.01045","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-02T09:47:40Z","cross_cats_sorted":[],"title_canon_sha256":"0d4fb0b0c1eca6497debcd8f079e31467e39a5a66410ceb95340b0797facdb1d","abstract_canon_sha256":"490bc3f32088d4624d7c113720c9add78b365af3ab6fa4c2407af354d2ea9b17"},"schema_version":"1.0"},"canonical_sha256":"20f51c17dd818b388fdec98a4beb91ca5108f364cce15a0175022ab02dc2b0a5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:43:56.241231Z","signature_b64":"YDWAdqGstElNAz9oS6ayFMIC+VZfBgzMUy6vuDA+uw6dnrnPUwsByg+ZmgpT0XEv5L197SWMuBzoT6pL+OlpBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20f51c17dd818b388fdec98a4beb91ca5108f364cce15a0175022ab02dc2b0a5","last_reissued_at":"2026-07-05T07:43:56.240723Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:43:56.240723Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2310.01045","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:43:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MLCjYVHz16UUh2EcX+nHdtBMUlYuOHM7Etv2iMh2Dp0Mt/Ze9PVjSC0DUlPmA4J4VZpHE6ApKwKJyLOGDp6+Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:51:01.578351Z"},"content_sha256":"9d4d6fc6ec52ab93cf1793a86ca8ccdff5bac5c3c268ff8f2c154128cc7813ed","schema_version":"1.0","event_id":"sha256:9d4d6fc6ec52ab93cf1793a86ca8ccdff5bac5c3c268ff8f2c154128cc7813ed"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:ED2RYF65QGFTRD66ZGFEX24RZJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Tool-Augmented Reward Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hao Tian, Hua Wu, Lei Li, Ningyu Zhang, Shuohuan Wang, Yekun Chai, Yu Sun","submitted_at":"2023-10-02T09:47:40Z","abstract_excerpt":"Reward modeling (a.k.a., preference modeling) is instrumental for aligning large language models with human preferences, particularly within the context of reinforcement learning from human feedback (RLHF). While conventional reward models (RMs) have exhibited remarkable scalability, they oft struggle with fundamental functionality such as arithmetic computation, code execution, and factual lookup. In this paper, we propose a tool-augmented preference modeling approach, named Themis, to address these limitations by empowering RMs with access to external environments, including calculators and "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.01045","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/2310.01045/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:43:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zvL6h6sd9n+8P3ulVOkyShpP5mgUuiD8Kj8l2GTCjHVyB5fEo3SXbsICRn9AnKqWyfE3aZQoLbWQZNr0I8KhAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T15:51:01.578732Z"},"content_sha256":"8984978d61f4aa84736b165697e5d452e5ea18011b8ad967fc1bfd37257ab36b","schema_version":"1.0","event_id":"sha256:8984978d61f4aa84736b165697e5d452e5ea18011b8ad967fc1bfd37257ab36b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ED2RYF65QGFTRD66ZGFEX24RZJ/bundle.json","state_url":"https://pith.science/pith/ED2RYF65QGFTRD66ZGFEX24RZJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ED2RYF65QGFTRD66ZGFEX24RZJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-06T15:51:01Z","links":{"resolver":"https://pith.science/pith/ED2RYF65QGFTRD66ZGFEX24RZJ","bundle":"https://pith.science/pith/ED2RYF65QGFTRD66ZGFEX24RZJ/bundle.json","state":"https://pith.science/pith/ED2RYF65QGFTRD66ZGFEX24RZJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ED2RYF65QGFTRD66ZGFEX24RZJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:ED2RYF65QGFTRD66ZGFEX24RZJ","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":"490bc3f32088d4624d7c113720c9add78b365af3ab6fa4c2407af354d2ea9b17","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-02T09:47:40Z","title_canon_sha256":"0d4fb0b0c1eca6497debcd8f079e31467e39a5a66410ceb95340b0797facdb1d"},"schema_version":"1.0","source":{"id":"2310.01045","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2310.01045","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"arxiv_version","alias_value":"2310.01045v2","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.01045","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"pith_short_12","alias_value":"ED2RYF65QGFT","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"pith_short_16","alias_value":"ED2RYF65QGFTRD66","created_at":"2026-07-05T07:43:56Z"},{"alias_kind":"pith_short_8","alias_value":"ED2RYF65","created_at":"2026-07-05T07:43:56Z"}],"graph_snapshots":[{"event_id":"sha256:8984978d61f4aa84736b165697e5d452e5ea18011b8ad967fc1bfd37257ab36b","target":"graph","created_at":"2026-07-05T07:43:56Z","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/2310.01045/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reward modeling (a.k.a., preference modeling) is instrumental for aligning large language models with human preferences, particularly within the context of reinforcement learning from human feedback (RLHF). While conventional reward models (RMs) have exhibited remarkable scalability, they oft struggle with fundamental functionality such as arithmetic computation, code execution, and factual lookup. In this paper, we propose a tool-augmented preference modeling approach, named Themis, to address these limitations by empowering RMs with access to external environments, including calculators and ","authors_text":"Hao Tian, Hua Wu, Lei Li, Ningyu Zhang, Shuohuan Wang, Yekun Chai, Yu Sun","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-02T09:47:40Z","title":"Tool-Augmented Reward Modeling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.01045","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":"sha256:9d4d6fc6ec52ab93cf1793a86ca8ccdff5bac5c3c268ff8f2c154128cc7813ed","target":"record","created_at":"2026-07-05T07:43:56Z","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":"490bc3f32088d4624d7c113720c9add78b365af3ab6fa4c2407af354d2ea9b17","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-02T09:47:40Z","title_canon_sha256":"0d4fb0b0c1eca6497debcd8f079e31467e39a5a66410ceb95340b0797facdb1d"},"schema_version":"1.0","source":{"id":"2310.01045","kind":"arxiv","version":2}},"canonical_sha256":"20f51c17dd818b388fdec98a4beb91ca5108f364cce15a0175022ab02dc2b0a5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"20f51c17dd818b388fdec98a4beb91ca5108f364cce15a0175022ab02dc2b0a5","first_computed_at":"2026-07-05T07:43:56.240723Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:43:56.240723Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YDWAdqGstElNAz9oS6ayFMIC+VZfBgzMUy6vuDA+uw6dnrnPUwsByg+ZmgpT0XEv5L197SWMuBzoT6pL+OlpBg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:43:56.241231Z","signed_message":"canonical_sha256_bytes"},"source_id":"2310.01045","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9d4d6fc6ec52ab93cf1793a86ca8ccdff5bac5c3c268ff8f2c154128cc7813ed","sha256:8984978d61f4aa84736b165697e5d452e5ea18011b8ad967fc1bfd37257ab36b"],"state_sha256":"7318cde792fd67c318fef5a681cad40668426f3d3b7b71f1793ba1af1baebd72"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"s405sNhFMaQw4U7zpeF9UMo1giqyxNOoR2WV/a7Yos3YGwlBVbyW4ExcseLUEP+F1u1Z3ZsB88XUf9P8o9PhCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T15:51:01.580660Z","bundle_sha256":"b48a61da2afe172f2030a14728359092a45825a8546d9728831c8c002788ffe3"}}