{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XKJ7LBJLMCDUV5JUAAVWBH37HD","short_pith_number":"pith:XKJ7LBJL","canonical_record":{"source":{"id":"2506.23235","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-29T13:45:54Z","cross_cats_sorted":[],"title_canon_sha256":"f9e7d976229a002655fef67026e1fa9dad997ff4cd3cbb6be7de8a8c1b4e0e72","abstract_canon_sha256":"b565b0b3aa984ea3955cb8b2bd2ddb962dc54ddad1b7b85e473be37ad56cbd50"},"schema_version":"1.0"},"canonical_sha256":"ba93f5852b60874af534002b609f7f38fca4c14b5b65404245aed8d5142e7adb","source":{"kind":"arxiv","id":"2506.23235","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.23235","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"arxiv_version","alias_value":"2506.23235v1","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.23235","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"pith_short_12","alias_value":"XKJ7LBJLMCDU","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"pith_short_16","alias_value":"XKJ7LBJLMCDUV5JU","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"pith_short_8","alias_value":"XKJ7LBJL","created_at":"2026-07-05T11:29:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XKJ7LBJLMCDUV5JUAAVWBH37HD","target":"record","payload":{"canonical_record":{"source":{"id":"2506.23235","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-29T13:45:54Z","cross_cats_sorted":[],"title_canon_sha256":"f9e7d976229a002655fef67026e1fa9dad997ff4cd3cbb6be7de8a8c1b4e0e72","abstract_canon_sha256":"b565b0b3aa984ea3955cb8b2bd2ddb962dc54ddad1b7b85e473be37ad56cbd50"},"schema_version":"1.0"},"canonical_sha256":"ba93f5852b60874af534002b609f7f38fca4c14b5b65404245aed8d5142e7adb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:29:18.334829Z","signature_b64":"RAFW4lLsXLr65m20JKNB57W2LXcKXAn6ufx2EuNBbS2GvO8vds+WWkqyzRRCpw7zyIY641Y5kuMTt6qgjhOVDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ba93f5852b60874af534002b609f7f38fca4c14b5b65404245aed8d5142e7adb","last_reissued_at":"2026-07-05T11:29:18.334303Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:29:18.334303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.23235","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:29:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KwKDdrogw+v7VdZBhJ/8+0Fjs0bK6QVqMLH+5xQFd4myQYWttKCVvSms0pbJuYJ3jbHTg3hbMXs+iQsurT4YCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T21:55:15.029239Z"},"content_sha256":"ec2d46bc05bf29edaf231080ae9c2f70b70f0c119a92e3962c468f31bf2fc3c2","schema_version":"1.0","event_id":"sha256:ec2d46bc05bf29edaf231080ae9c2f70b70f0c119a92e3962c468f31bf2fc3c2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XKJ7LBJLMCDUV5JUAAVWBH37HD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Generalist Reward Models: Found Inside Large Language Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Lei Yuan, Ningjing Chao, Tian Xu, Xiong-Hui Chen, Xuqin Zhang, Yang Yu, Yi-Chen Li, Zhi-Hua Zhou, Zhongxiang Ling","submitted_at":"2025-06-29T13:45:54Z","abstract_excerpt":"The alignment of Large Language Models (LLMs) is critically dependent on reward models trained on costly human preference data. While recent work explores bypassing this cost with AI feedback, these methods often lack a rigorous theoretical foundation. In this paper, we discover that a powerful generalist reward model is already latently present within any LLM trained via standard next-token prediction. We prove that this endogenous reward is not a heuristic, but is theoretically equivalent to a reward function learned through offline inverse reinforcement learning. This connection allows us t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.23235","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/2506.23235/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":"2026-07-05T11:29:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RCxSCJx4/olOcI3Ew9Nat2pyyJCHnLc1soZd0rmSNlV50NiRTV/PJLRnwwWXCjfouaP7hbKuqRZbFeb6wLa0DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T21:55:15.030463Z"},"content_sha256":"386029ef5b91d563358329d6c52155500a90c1a55479e5f7d18d34327f3324f1","schema_version":"1.0","event_id":"sha256:386029ef5b91d563358329d6c52155500a90c1a55479e5f7d18d34327f3324f1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XKJ7LBJLMCDUV5JUAAVWBH37HD/bundle.json","state_url":"https://pith.science/pith/XKJ7LBJLMCDUV5JUAAVWBH37HD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XKJ7LBJLMCDUV5JUAAVWBH37HD/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":"2026-08-06T21:55:15Z","links":{"resolver":"https://pith.science/pith/XKJ7LBJLMCDUV5JUAAVWBH37HD","bundle":"https://pith.science/pith/XKJ7LBJLMCDUV5JUAAVWBH37HD/bundle.json","state":"https://pith.science/pith/XKJ7LBJLMCDUV5JUAAVWBH37HD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XKJ7LBJLMCDUV5JUAAVWBH37HD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XKJ7LBJLMCDUV5JUAAVWBH37HD","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":"b565b0b3aa984ea3955cb8b2bd2ddb962dc54ddad1b7b85e473be37ad56cbd50","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-29T13:45:54Z","title_canon_sha256":"f9e7d976229a002655fef67026e1fa9dad997ff4cd3cbb6be7de8a8c1b4e0e72"},"schema_version":"1.0","source":{"id":"2506.23235","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.23235","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"arxiv_version","alias_value":"2506.23235v1","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.23235","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"pith_short_12","alias_value":"XKJ7LBJLMCDU","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"pith_short_16","alias_value":"XKJ7LBJLMCDUV5JU","created_at":"2026-07-05T11:29:18Z"},{"alias_kind":"pith_short_8","alias_value":"XKJ7LBJL","created_at":"2026-07-05T11:29:18Z"}],"graph_snapshots":[{"event_id":"sha256:386029ef5b91d563358329d6c52155500a90c1a55479e5f7d18d34327f3324f1","target":"graph","created_at":"2026-07-05T11:29:18Z","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/2506.23235/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The alignment of Large Language Models (LLMs) is critically dependent on reward models trained on costly human preference data. While recent work explores bypassing this cost with AI feedback, these methods often lack a rigorous theoretical foundation. In this paper, we discover that a powerful generalist reward model is already latently present within any LLM trained via standard next-token prediction. We prove that this endogenous reward is not a heuristic, but is theoretically equivalent to a reward function learned through offline inverse reinforcement learning. This connection allows us t","authors_text":"Lei Yuan, Ningjing Chao, Tian Xu, Xiong-Hui Chen, Xuqin Zhang, Yang Yu, Yi-Chen Li, Zhi-Hua Zhou, Zhongxiang Ling","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-29T13:45:54Z","title":"Generalist Reward Models: Found Inside Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.23235","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:ec2d46bc05bf29edaf231080ae9c2f70b70f0c119a92e3962c468f31bf2fc3c2","target":"record","created_at":"2026-07-05T11:29:18Z","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":"b565b0b3aa984ea3955cb8b2bd2ddb962dc54ddad1b7b85e473be37ad56cbd50","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-29T13:45:54Z","title_canon_sha256":"f9e7d976229a002655fef67026e1fa9dad997ff4cd3cbb6be7de8a8c1b4e0e72"},"schema_version":"1.0","source":{"id":"2506.23235","kind":"arxiv","version":1}},"canonical_sha256":"ba93f5852b60874af534002b609f7f38fca4c14b5b65404245aed8d5142e7adb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ba93f5852b60874af534002b609f7f38fca4c14b5b65404245aed8d5142e7adb","first_computed_at":"2026-07-05T11:29:18.334303Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:29:18.334303Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"RAFW4lLsXLr65m20JKNB57W2LXcKXAn6ufx2EuNBbS2GvO8vds+WWkqyzRRCpw7zyIY641Y5kuMTt6qgjhOVDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:29:18.334829Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.23235","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ec2d46bc05bf29edaf231080ae9c2f70b70f0c119a92e3962c468f31bf2fc3c2","sha256:386029ef5b91d563358329d6c52155500a90c1a55479e5f7d18d34327f3324f1"],"state_sha256":"d5bbf02a089262ce007a5cb0b7118b91c87796601037f01e56b66b01a11b2ae1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w38hwqjTpSgwjbtI7rCG/b3G18UrTt1NdlJFP9gNr4QqoWjm5gUMzP2Btj/2DR4aGKdn1jYOqgs723gk73CHAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T21:55:15.036888Z","bundle_sha256":"ab46bdfb0033a8b3eebc59f2c761eb958ce25982fd03bf8a476bc25783d0cc5b"}}