{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:LBLXHQPSO7GEYO6VN2BYR2YQDL","short_pith_number":"pith:LBLXHQPS","canonical_record":{"source":{"id":"2312.07401","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-12T16:17:15Z","cross_cats_sorted":[],"title_canon_sha256":"154dbe55ee0f5e8b13f7053a1c1e73f073752fe28d629c75fa65e3aa862c3758","abstract_canon_sha256":"20da1564e0b8a196cb9897263f2dacd2567cd1a74afbe45a8d5b425ec6cb2e7e"},"schema_version":"1.0"},"canonical_sha256":"585773c1f277cc4c3bd56e8388eb101ac9a9118bef3e09b88868600e4c550f1f","source":{"kind":"arxiv","id":"2312.07401","version":5},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.07401","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"arxiv_version","alias_value":"2312.07401v5","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.07401","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"pith_short_12","alias_value":"LBLXHQPSO7GE","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"pith_short_16","alias_value":"LBLXHQPSO7GEYO6V","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"pith_short_8","alias_value":"LBLXHQPS","created_at":"2026-07-05T09:16:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:LBLXHQPSO7GEYO6VN2BYR2YQDL","target":"record","payload":{"canonical_record":{"source":{"id":"2312.07401","kind":"arxiv","version":5},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-12T16:17:15Z","cross_cats_sorted":[],"title_canon_sha256":"154dbe55ee0f5e8b13f7053a1c1e73f073752fe28d629c75fa65e3aa862c3758","abstract_canon_sha256":"20da1564e0b8a196cb9897263f2dacd2567cd1a74afbe45a8d5b425ec6cb2e7e"},"schema_version":"1.0"},"canonical_sha256":"585773c1f277cc4c3bd56e8388eb101ac9a9118bef3e09b88868600e4c550f1f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:10.685162Z","signature_b64":"9EvolLG4074s+ZP8bq3iJ+/lynbEXyhVMKYa1djSxElzwLL2/wqya48KiGYskGqxoylHUfJhfeWzkCLvZJxHAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"585773c1f277cc4c3bd56e8388eb101ac9a9118bef3e09b88868600e4c550f1f","last_reissued_at":"2026-07-05T09:16:10.684669Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:10.684669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2312.07401","source_version":5,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:16:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"b6dwMKkiGtnHd5K2piT/ZvVV37T9/XZICyNM0Oq1koG2CMnLystviNz6v0xNR7oA2nwIrVmz77ayAStZ8d0IBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T00:58:12.561453Z"},"content_sha256":"7ff29fa4e00a3461d9c9cda711cb8d1e187ce822aaa5c078ff69ddb748f71717","schema_version":"1.0","event_id":"sha256:7ff29fa4e00a3461d9c9cda711cb8d1e187ce822aaa5c078ff69ddb748f71717"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:LBLXHQPSO7GEYO6VN2BYR2YQDL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On Diversified Preferences of Large Language Model Alignment","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Dun Zeng, Longyue Wang, Nan Du, Pengyu Cheng, Tianhao Hu, Wanshun Chen, Yong Dai, Zenglin Xu","submitted_at":"2023-12-12T16:17:15Z","abstract_excerpt":"Aligning large language models (LLMs) with human preferences has been recognized as the key to improving LLMs' interaction quality. However, in this pluralistic world, human preferences can be diversified due to annotators' different tastes, which hinders the effectiveness of LLM alignment methods. This paper presents the first quantitative analysis of the experimental scaling law for reward models with varying sizes, from 1.3 billion to 7 billion parameters, trained with human feedback exhibiting diverse preferences. Our analysis reveals that the impact of diversified human preferences depend"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.07401","kind":"arxiv","version":5},"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/2312.07401/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":"2026-07-05T09:16:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"z2c6Ofl4D6mK5fAYeePnkS+rfX082UbbeQf7A3WI6Zi/5km7rsukyeiQdBHRiq6uGFIwA5TYqI3zq9f5A+EnCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T00:58:12.562024Z"},"content_sha256":"72fe5f698582946986f451ce83e5743225d057bc5c6f18456afdcb567f5fd7f8","schema_version":"1.0","event_id":"sha256:72fe5f698582946986f451ce83e5743225d057bc5c6f18456afdcb567f5fd7f8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LBLXHQPSO7GEYO6VN2BYR2YQDL/bundle.json","state_url":"https://pith.science/pith/LBLXHQPSO7GEYO6VN2BYR2YQDL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LBLXHQPSO7GEYO6VN2BYR2YQDL/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":"2026-08-10T00:58:12Z","links":{"resolver":"https://pith.science/pith/LBLXHQPSO7GEYO6VN2BYR2YQDL","bundle":"https://pith.science/pith/LBLXHQPSO7GEYO6VN2BYR2YQDL/bundle.json","state":"https://pith.science/pith/LBLXHQPSO7GEYO6VN2BYR2YQDL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LBLXHQPSO7GEYO6VN2BYR2YQDL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:LBLXHQPSO7GEYO6VN2BYR2YQDL","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":"20da1564e0b8a196cb9897263f2dacd2567cd1a74afbe45a8d5b425ec6cb2e7e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-12T16:17:15Z","title_canon_sha256":"154dbe55ee0f5e8b13f7053a1c1e73f073752fe28d629c75fa65e3aa862c3758"},"schema_version":"1.0","source":{"id":"2312.07401","kind":"arxiv","version":5}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.07401","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"arxiv_version","alias_value":"2312.07401v5","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.07401","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"pith_short_12","alias_value":"LBLXHQPSO7GE","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"pith_short_16","alias_value":"LBLXHQPSO7GEYO6V","created_at":"2026-07-05T09:16:10Z"},{"alias_kind":"pith_short_8","alias_value":"LBLXHQPS","created_at":"2026-07-05T09:16:10Z"}],"graph_snapshots":[{"event_id":"sha256:72fe5f698582946986f451ce83e5743225d057bc5c6f18456afdcb567f5fd7f8","target":"graph","created_at":"2026-07-05T09:16:10Z","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/2312.07401/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Aligning large language models (LLMs) with human preferences has been recognized as the key to improving LLMs' interaction quality. However, in this pluralistic world, human preferences can be diversified due to annotators' different tastes, which hinders the effectiveness of LLM alignment methods. This paper presents the first quantitative analysis of the experimental scaling law for reward models with varying sizes, from 1.3 billion to 7 billion parameters, trained with human feedback exhibiting diverse preferences. Our analysis reveals that the impact of diversified human preferences depend","authors_text":"Dun Zeng, Longyue Wang, Nan Du, Pengyu Cheng, Tianhao Hu, Wanshun Chen, Yong Dai, Zenglin Xu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-12T16:17:15Z","title":"On Diversified Preferences of Large Language Model Alignment"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.07401","kind":"arxiv","version":5},"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":"sha256:7ff29fa4e00a3461d9c9cda711cb8d1e187ce822aaa5c078ff69ddb748f71717","target":"record","created_at":"2026-07-05T09:16:10Z","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":"20da1564e0b8a196cb9897263f2dacd2567cd1a74afbe45a8d5b425ec6cb2e7e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2023-12-12T16:17:15Z","title_canon_sha256":"154dbe55ee0f5e8b13f7053a1c1e73f073752fe28d629c75fa65e3aa862c3758"},"schema_version":"1.0","source":{"id":"2312.07401","kind":"arxiv","version":5}},"canonical_sha256":"585773c1f277cc4c3bd56e8388eb101ac9a9118bef3e09b88868600e4c550f1f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"585773c1f277cc4c3bd56e8388eb101ac9a9118bef3e09b88868600e4c550f1f","first_computed_at":"2026-07-05T09:16:10.684669Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:16:10.684669Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9EvolLG4074s+ZP8bq3iJ+/lynbEXyhVMKYa1djSxElzwLL2/wqya48KiGYskGqxoylHUfJhfeWzkCLvZJxHAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:16:10.685162Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.07401","source_kind":"arxiv","source_version":5}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7ff29fa4e00a3461d9c9cda711cb8d1e187ce822aaa5c078ff69ddb748f71717","sha256:72fe5f698582946986f451ce83e5743225d057bc5c6f18456afdcb567f5fd7f8"],"state_sha256":"eb5485f2f133d8e05725db01fe20b832c83400d2a57dc370bd55388e3dcd3722"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iGIx8R+0SCFxDwUTB9DUe5fCGyKPkhWOKUHReiuKTPp+gIPetTNZAeSZ28fshwCQINLl/QJGf7GgfPO3UolGDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T00:58:12.567155Z","bundle_sha256":"c394344f09be55ee309a1e3cd05c69b05c94f50ee857f9489d2642e160331fa3"}}