{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:B7T6I6XL4XT43BD7FJX2TXEJF6","short_pith_number":"pith:B7T6I6XL","canonical_record":{"source":{"id":"2505.23927","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-29T18:22:02Z","cross_cats_sorted":[],"title_canon_sha256":"84645f304f59984b80befa9139272fa1a7a63327a2623482a059795dd760a477","abstract_canon_sha256":"2cfb135eeb8fa37b38a0133397b0f181dee8b56f43268ed9e7422b1f7b8ad8e9"},"schema_version":"1.0"},"canonical_sha256":"0fe7e47aebe5e7cd847f2a6fa9dc892f96f996f3e78320bb3df5620eb96cbec5","source":{"kind":"arxiv","id":"2505.23927","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.23927","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"arxiv_version","alias_value":"2505.23927v1","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23927","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"pith_short_12","alias_value":"B7T6I6XL4XT4","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"pith_short_16","alias_value":"B7T6I6XL4XT43BD7","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"pith_short_8","alias_value":"B7T6I6XL","created_at":"2026-07-05T11:12:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:B7T6I6XL4XT43BD7FJX2TXEJF6","target":"record","payload":{"canonical_record":{"source":{"id":"2505.23927","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-29T18:22:02Z","cross_cats_sorted":[],"title_canon_sha256":"84645f304f59984b80befa9139272fa1a7a63327a2623482a059795dd760a477","abstract_canon_sha256":"2cfb135eeb8fa37b38a0133397b0f181dee8b56f43268ed9e7422b1f7b8ad8e9"},"schema_version":"1.0"},"canonical_sha256":"0fe7e47aebe5e7cd847f2a6fa9dc892f96f996f3e78320bb3df5620eb96cbec5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:37.946984Z","signature_b64":"Don/nHh+nRaVUyCUN1rJ9EPoLRu/yV0vSCzYRM/C1AuGL6JpfhbnOqctT4yZ4HdXf7TVN1qxVKWgsbKQL0FkBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0fe7e47aebe5e7cd847f2a6fa9dc892f96f996f3e78320bb3df5620eb96cbec5","last_reissued_at":"2026-07-05T11:12:37.946486Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:37.946486Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.23927","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":"2026-07-05T11:12:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4YtlRh6Ek2X7QBtN5OjSAnq6eVS/XLzfmac0YRs8H7bwRqaCY1OfM4DNUxDIxw1n53XicWYNF1Y1DLUy1x7qCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T18:36:20.869331Z"},"content_sha256":"36e6b81f9444aead4093a8d0fc91992ab52d5136ca92120af19156a39eb2e7db","schema_version":"1.0","event_id":"sha256:36e6b81f9444aead4093a8d0fc91992ab52d5136ca92120af19156a39eb2e7db"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:B7T6I6XL4XT43BD7FJX2TXEJF6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Thompson Sampling in Online RLHF with General Function Approximation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Jie Fu, Songtao Feng","submitted_at":"2025-05-29T18:22:02Z","abstract_excerpt":"Reinforcement learning from human feedback (RLHF) has achieved great empirical success in aligning large language models (LLMs) with human preference, and it is of great importance to study the statistical efficiency of RLHF algorithms from a theoretical perspective. In this work, we consider the online RLHF setting where the preference data is revealed during the learning process and study action value function approximation. We design a model-free posterior sampling algorithm for online RLHF inspired by Thompson sampling and provide its theoretical guarantee. Specifically, we adopt Bellman e"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23927","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/2505.23927/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":"2026-07-05T11:12:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2vS8NPxQ5ZVvj/qfysYnAdXHED+GZpxiXXZzXpC/SV4cmUAFubleZ6ZoHvEIny7C8sSbSiLul9dv69o5UbxuDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T18:36:20.869853Z"},"content_sha256":"d7803a3723db4862decfbf634d7a393844ba75bd2216ba7aff07dc73d4f2d6c9","schema_version":"1.0","event_id":"sha256:d7803a3723db4862decfbf634d7a393844ba75bd2216ba7aff07dc73d4f2d6c9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/B7T6I6XL4XT43BD7FJX2TXEJF6/bundle.json","state_url":"https://pith.science/pith/B7T6I6XL4XT43BD7FJX2TXEJF6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/B7T6I6XL4XT43BD7FJX2TXEJF6/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":"2026-08-09T18:36:20Z","links":{"resolver":"https://pith.science/pith/B7T6I6XL4XT43BD7FJX2TXEJF6","bundle":"https://pith.science/pith/B7T6I6XL4XT43BD7FJX2TXEJF6/bundle.json","state":"https://pith.science/pith/B7T6I6XL4XT43BD7FJX2TXEJF6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/B7T6I6XL4XT43BD7FJX2TXEJF6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:B7T6I6XL4XT43BD7FJX2TXEJF6","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":"2cfb135eeb8fa37b38a0133397b0f181dee8b56f43268ed9e7422b1f7b8ad8e9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-29T18:22:02Z","title_canon_sha256":"84645f304f59984b80befa9139272fa1a7a63327a2623482a059795dd760a477"},"schema_version":"1.0","source":{"id":"2505.23927","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.23927","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"arxiv_version","alias_value":"2505.23927v1","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23927","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"pith_short_12","alias_value":"B7T6I6XL4XT4","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"pith_short_16","alias_value":"B7T6I6XL4XT43BD7","created_at":"2026-07-05T11:12:37Z"},{"alias_kind":"pith_short_8","alias_value":"B7T6I6XL","created_at":"2026-07-05T11:12:37Z"}],"graph_snapshots":[{"event_id":"sha256:d7803a3723db4862decfbf634d7a393844ba75bd2216ba7aff07dc73d4f2d6c9","target":"graph","created_at":"2026-07-05T11:12:37Z","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/2505.23927/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning from human feedback (RLHF) has achieved great empirical success in aligning large language models (LLMs) with human preference, and it is of great importance to study the statistical efficiency of RLHF algorithms from a theoretical perspective. In this work, we consider the online RLHF setting where the preference data is revealed during the learning process and study action value function approximation. We design a model-free posterior sampling algorithm for online RLHF inspired by Thompson sampling and provide its theoretical guarantee. Specifically, we adopt Bellman e","authors_text":"Jie Fu, Songtao Feng","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-29T18:22:02Z","title":"Thompson Sampling in Online RLHF with General Function Approximation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23927","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":"sha256:36e6b81f9444aead4093a8d0fc91992ab52d5136ca92120af19156a39eb2e7db","target":"record","created_at":"2026-07-05T11:12:37Z","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":"2cfb135eeb8fa37b38a0133397b0f181dee8b56f43268ed9e7422b1f7b8ad8e9","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-29T18:22:02Z","title_canon_sha256":"84645f304f59984b80befa9139272fa1a7a63327a2623482a059795dd760a477"},"schema_version":"1.0","source":{"id":"2505.23927","kind":"arxiv","version":1}},"canonical_sha256":"0fe7e47aebe5e7cd847f2a6fa9dc892f96f996f3e78320bb3df5620eb96cbec5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0fe7e47aebe5e7cd847f2a6fa9dc892f96f996f3e78320bb3df5620eb96cbec5","first_computed_at":"2026-07-05T11:12:37.946486Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:12:37.946486Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Don/nHh+nRaVUyCUN1rJ9EPoLRu/yV0vSCzYRM/C1AuGL6JpfhbnOqctT4yZ4HdXf7TVN1qxVKWgsbKQL0FkBw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:12:37.946984Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.23927","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:36e6b81f9444aead4093a8d0fc91992ab52d5136ca92120af19156a39eb2e7db","sha256:d7803a3723db4862decfbf634d7a393844ba75bd2216ba7aff07dc73d4f2d6c9"],"state_sha256":"008611324be6d695f002c3e00d0775c4873fa1f9eaef22046eb0d0be76b3a3e0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fQIe6KnTXGZP6aW0ZfSyVKp7MfTJeN6cTOBFl9UJ1l89yZLs4UDLIOKb9ewWScoxUMjh2P7Fb9EXBsrlYdFnDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T18:36:20.873856Z","bundle_sha256":"629f9c51feb4dc0de4bb68c9fba69b07295b96d4f6da08208f53b410677b483e"}}