{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:6XBWKF7W6XIEM4FE4ZUZCIGZQP","short_pith_number":"pith:6XBWKF7W","canonical_record":{"source":{"id":"2309.15028","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-26T15:57:57Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"fa2c3d47e4d98188a681c7941754b968604429b373785ce293c0a1b70179a920","abstract_canon_sha256":"9dec11925715ef63b6ad2021ae34cc615e43424dbc5cc97d38022f37914c3245"},"schema_version":"1.0"},"canonical_sha256":"f5c36517f6f5d04670a4e6699120d983ca6ce8c08f4dd93e066763dad55f37a8","source":{"kind":"arxiv","id":"2309.15028","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.15028","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"arxiv_version","alias_value":"2309.15028v3","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.15028","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"pith_short_12","alias_value":"6XBWKF7W6XIE","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"pith_short_16","alias_value":"6XBWKF7W6XIEM4FE","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"pith_short_8","alias_value":"6XBWKF7W","created_at":"2026-07-05T08:03:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:6XBWKF7W6XIEM4FE4ZUZCIGZQP","target":"record","payload":{"canonical_record":{"source":{"id":"2309.15028","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-26T15:57:57Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"fa2c3d47e4d98188a681c7941754b968604429b373785ce293c0a1b70179a920","abstract_canon_sha256":"9dec11925715ef63b6ad2021ae34cc615e43424dbc5cc97d38022f37914c3245"},"schema_version":"1.0"},"canonical_sha256":"f5c36517f6f5d04670a4e6699120d983ca6ce8c08f4dd93e066763dad55f37a8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:03:26.717440Z","signature_b64":"GCdRPqp/s3OkLgWGSE93XtsCberhqCkjpaa9NF0s+szfWAzQ7y/FQZrThFs9TJ4E579m53SBsAjgk4vR/i/YDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f5c36517f6f5d04670a4e6699120d983ca6ce8c08f4dd93e066763dad55f37a8","last_reissued_at":"2026-07-05T08:03:26.716837Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:03:26.716837Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2309.15028","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:03:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zrKZFSp1kT3d/YW3oSOckptzA0TKPSuZOUO9N/gx8UPNJcrq8AoGTLMUqeMIN3s9gfUUhdTPV/WDB+MX91wTCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T22:56:29.536574Z"},"content_sha256":"93443df19e2d5f4e88788f81790bcfcf3b46904c37029054e1eeb09843fca680","schema_version":"1.0","event_id":"sha256:93443df19e2d5f4e88788f81790bcfcf3b46904c37029054e1eeb09843fca680"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:6XBWKF7W6XIEM4FE4ZUZCIGZQP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Don't throw away your value model! Generating more preferable text with Value-Guided Monte-Carlo Tree Search decoding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Andrew Cohen, Asli Celikyilmaz, Hannaneh Hajishirzi, Jiacheng Liu, Ramakanth Pasunuru, Yejin Choi","submitted_at":"2023-09-26T15:57:57Z","abstract_excerpt":"Inference-time search algorithms such as Monte-Carlo Tree Search (MCTS) may seem unnecessary when generating natural language text based on state-of-the-art reinforcement learning such as Proximal Policy Optimization (PPO). In this paper, we demonstrate that it is possible to get extra mileage out of PPO by integrating MCTS on top. The key idea is not to throw out the value network, a byproduct of PPO training for evaluating partial output sequences, when decoding text out of the policy network. More concretely, we present a novel value-guided decoding algorithm called PPO-MCTS, which can inte"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.15028","kind":"arxiv","version":3},"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/2309.15028/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":"2026-07-05T08:03:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BWLNbAQQAviRSThE9nkymLXM1y7x6x28mJXyMqpVPavgbTwc4ySeDmRHtkykng2IzwFywnzE/9O2sYGN/xXhAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T22:56:29.537591Z"},"content_sha256":"83056578db56391dfecc799735d00256e91b9805ee6d22d13e40456fde33a362","schema_version":"1.0","event_id":"sha256:83056578db56391dfecc799735d00256e91b9805ee6d22d13e40456fde33a362"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6XBWKF7W6XIEM4FE4ZUZCIGZQP/bundle.json","state_url":"https://pith.science/pith/6XBWKF7W6XIEM4FE4ZUZCIGZQP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6XBWKF7W6XIEM4FE4ZUZCIGZQP/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":"2026-08-06T22:56:29Z","links":{"resolver":"https://pith.science/pith/6XBWKF7W6XIEM4FE4ZUZCIGZQP","bundle":"https://pith.science/pith/6XBWKF7W6XIEM4FE4ZUZCIGZQP/bundle.json","state":"https://pith.science/pith/6XBWKF7W6XIEM4FE4ZUZCIGZQP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6XBWKF7W6XIEM4FE4ZUZCIGZQP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:6XBWKF7W6XIEM4FE4ZUZCIGZQP","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":"9dec11925715ef63b6ad2021ae34cc615e43424dbc5cc97d38022f37914c3245","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-26T15:57:57Z","title_canon_sha256":"fa2c3d47e4d98188a681c7941754b968604429b373785ce293c0a1b70179a920"},"schema_version":"1.0","source":{"id":"2309.15028","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.15028","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"arxiv_version","alias_value":"2309.15028v3","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.15028","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"pith_short_12","alias_value":"6XBWKF7W6XIE","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"pith_short_16","alias_value":"6XBWKF7W6XIEM4FE","created_at":"2026-07-05T08:03:26Z"},{"alias_kind":"pith_short_8","alias_value":"6XBWKF7W","created_at":"2026-07-05T08:03:26Z"}],"graph_snapshots":[{"event_id":"sha256:83056578db56391dfecc799735d00256e91b9805ee6d22d13e40456fde33a362","target":"graph","created_at":"2026-07-05T08:03:26Z","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/2309.15028/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Inference-time search algorithms such as Monte-Carlo Tree Search (MCTS) may seem unnecessary when generating natural language text based on state-of-the-art reinforcement learning such as Proximal Policy Optimization (PPO). In this paper, we demonstrate that it is possible to get extra mileage out of PPO by integrating MCTS on top. The key idea is not to throw out the value network, a byproduct of PPO training for evaluating partial output sequences, when decoding text out of the policy network. More concretely, we present a novel value-guided decoding algorithm called PPO-MCTS, which can inte","authors_text":"Andrew Cohen, Asli Celikyilmaz, Hannaneh Hajishirzi, Jiacheng Liu, Ramakanth Pasunuru, Yejin Choi","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-26T15:57:57Z","title":"Don't throw away your value model! Generating more preferable text with Value-Guided Monte-Carlo Tree Search decoding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.15028","kind":"arxiv","version":3},"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":"sha256:93443df19e2d5f4e88788f81790bcfcf3b46904c37029054e1eeb09843fca680","target":"record","created_at":"2026-07-05T08:03:26Z","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":"9dec11925715ef63b6ad2021ae34cc615e43424dbc5cc97d38022f37914c3245","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-26T15:57:57Z","title_canon_sha256":"fa2c3d47e4d98188a681c7941754b968604429b373785ce293c0a1b70179a920"},"schema_version":"1.0","source":{"id":"2309.15028","kind":"arxiv","version":3}},"canonical_sha256":"f5c36517f6f5d04670a4e6699120d983ca6ce8c08f4dd93e066763dad55f37a8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f5c36517f6f5d04670a4e6699120d983ca6ce8c08f4dd93e066763dad55f37a8","first_computed_at":"2026-07-05T08:03:26.716837Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:03:26.716837Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GCdRPqp/s3OkLgWGSE93XtsCberhqCkjpaa9NF0s+szfWAzQ7y/FQZrThFs9TJ4E579m53SBsAjgk4vR/i/YDA==","signature_status":"signed_v1","signed_at":"2026-07-05T08:03:26.717440Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.15028","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:93443df19e2d5f4e88788f81790bcfcf3b46904c37029054e1eeb09843fca680","sha256:83056578db56391dfecc799735d00256e91b9805ee6d22d13e40456fde33a362"],"state_sha256":"10a80886081696364ee3cef342ac67f229e118144785a7a0db746202d5cf3ead"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PEOZawTi4fIYIrM44YuJt9ovD3bhVuIc5TvepK8sk+8xjsU5kKWyDxUof+99HgEp02zECeXwSfGYhvghPB2+Bw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T22:56:29.543215Z","bundle_sha256":"10c80e0807297a4a1d1dbd1ed7239afcafca189e16aba3d7daff8e2dbcfd100d"}}