{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:DCWWDQBVITJIZY43EADYBGEMIX","short_pith_number":"pith:DCWWDQBV","canonical_record":{"source":{"id":"2206.11871","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-06-05T18:38:42Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7bd07572af659ab055a87f1cc262e699f73bcbd46a3eb3027f53bce40d10ee33","abstract_canon_sha256":"6e6eb7c5cdc5d339c2338e86f390df664523e017d4bdc4c813b514ec43b561ff"},"schema_version":"1.0"},"canonical_sha256":"18ad61c03544d28ce39b200780988c45d910353bd75ab1744f22b768af813cd3","source":{"kind":"arxiv","id":"2206.11871","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2206.11871","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"arxiv_version","alias_value":"2206.11871v2","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.11871","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"pith_short_12","alias_value":"DCWWDQBVITJI","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"pith_short_16","alias_value":"DCWWDQBVITJIZY43","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"pith_short_8","alias_value":"DCWWDQBV","created_at":"2026-07-05T06:05:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:DCWWDQBVITJIZY43EADYBGEMIX","target":"record","payload":{"canonical_record":{"source":{"id":"2206.11871","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-06-05T18:38:42Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7bd07572af659ab055a87f1cc262e699f73bcbd46a3eb3027f53bce40d10ee33","abstract_canon_sha256":"6e6eb7c5cdc5d339c2338e86f390df664523e017d4bdc4c813b514ec43b561ff"},"schema_version":"1.0"},"canonical_sha256":"18ad61c03544d28ce39b200780988c45d910353bd75ab1744f22b768af813cd3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:05:27.571571Z","signature_b64":"NLTLNQdWyF6LFQ4ji2qoyDgRqh8GTVbw58Py4zpU7tE+FqdYxHM5X/HzskYc7Pf0R9x4QUZmPxW1xERXT50DBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"18ad61c03544d28ce39b200780988c45d910353bd75ab1744f22b768af813cd3","last_reissued_at":"2026-07-05T06:05:27.571080Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:05:27.571080Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2206.11871","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":"2026-07-05T06:05:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UcEmSZbHelgXm+ZtE7M3HVC1xbJ1h+F1E8Y8UHHwa8AmNdXIvvkgogb8+H9OWm9BH4JwvP0FZr1m2Hvh5VLoBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T12:58:53.296691Z"},"content_sha256":"79b66dfd46d972cbd1e9e08f383eb0af2330e3de4a981adee0d9501d1a351ee1","schema_version":"1.0","event_id":"sha256:79b66dfd46d972cbd1e9e08f383eb0af2330e3de4a981adee0d9501d1a351ee1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:DCWWDQBVITJIZY43EADYBGEMIX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Offline RL for Natural Language Generation with Implicit Language Q Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Charlie Snell, Ilya Kostrikov, Mengjiao Yang, Sergey Levine, Yi Su","submitted_at":"2022-06-05T18:38:42Z","abstract_excerpt":"Large language models distill broad knowledge from text corpora. However, they can be inconsistent when it comes to completing user specified tasks. This issue can be addressed by finetuning such models via supervised learning on curated datasets, or via reinforcement learning. In this work, we propose a novel offline RL method, implicit language Q-learning (ILQL), designed for use on language models, that combines both the flexible utility maximization framework of RL algorithms with the ability of supervised learning to leverage previously collected data, as well as its simplicity and stabil"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.11871","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/2206.11871/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":"2026-07-05T06:05:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LzuLpQHJEsvCKPoT0X9KCJODw+vbCKqE1COqckGP72q1G621o46y1hY5xTb2sNO/f/JZupKtuQyRaSyyA3epAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T12:58:53.297219Z"},"content_sha256":"4bbd3b3ec0b1b072e46a42a8a4f1bb01abef195876baa64070b4f4fa0d69a7ff","schema_version":"1.0","event_id":"sha256:4bbd3b3ec0b1b072e46a42a8a4f1bb01abef195876baa64070b4f4fa0d69a7ff"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DCWWDQBVITJIZY43EADYBGEMIX/bundle.json","state_url":"https://pith.science/pith/DCWWDQBVITJIZY43EADYBGEMIX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DCWWDQBVITJIZY43EADYBGEMIX/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":"2026-08-08T12:58:53Z","links":{"resolver":"https://pith.science/pith/DCWWDQBVITJIZY43EADYBGEMIX","bundle":"https://pith.science/pith/DCWWDQBVITJIZY43EADYBGEMIX/bundle.json","state":"https://pith.science/pith/DCWWDQBVITJIZY43EADYBGEMIX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DCWWDQBVITJIZY43EADYBGEMIX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:DCWWDQBVITJIZY43EADYBGEMIX","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":"6e6eb7c5cdc5d339c2338e86f390df664523e017d4bdc4c813b514ec43b561ff","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-06-05T18:38:42Z","title_canon_sha256":"7bd07572af659ab055a87f1cc262e699f73bcbd46a3eb3027f53bce40d10ee33"},"schema_version":"1.0","source":{"id":"2206.11871","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2206.11871","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"arxiv_version","alias_value":"2206.11871v2","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.11871","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"pith_short_12","alias_value":"DCWWDQBVITJI","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"pith_short_16","alias_value":"DCWWDQBVITJIZY43","created_at":"2026-07-05T06:05:27Z"},{"alias_kind":"pith_short_8","alias_value":"DCWWDQBV","created_at":"2026-07-05T06:05:27Z"}],"graph_snapshots":[{"event_id":"sha256:4bbd3b3ec0b1b072e46a42a8a4f1bb01abef195876baa64070b4f4fa0d69a7ff","target":"graph","created_at":"2026-07-05T06:05:27Z","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/2206.11871/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models distill broad knowledge from text corpora. However, they can be inconsistent when it comes to completing user specified tasks. This issue can be addressed by finetuning such models via supervised learning on curated datasets, or via reinforcement learning. In this work, we propose a novel offline RL method, implicit language Q-learning (ILQL), designed for use on language models, that combines both the flexible utility maximization framework of RL algorithms with the ability of supervised learning to leverage previously collected data, as well as its simplicity and stabil","authors_text":"Charlie Snell, Ilya Kostrikov, Mengjiao Yang, Sergey Levine, Yi Su","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-06-05T18:38:42Z","title":"Offline RL for Natural Language Generation with Implicit Language Q Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.11871","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":"sha256:79b66dfd46d972cbd1e9e08f383eb0af2330e3de4a981adee0d9501d1a351ee1","target":"record","created_at":"2026-07-05T06:05:27Z","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":"6e6eb7c5cdc5d339c2338e86f390df664523e017d4bdc4c813b514ec43b561ff","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-06-05T18:38:42Z","title_canon_sha256":"7bd07572af659ab055a87f1cc262e699f73bcbd46a3eb3027f53bce40d10ee33"},"schema_version":"1.0","source":{"id":"2206.11871","kind":"arxiv","version":2}},"canonical_sha256":"18ad61c03544d28ce39b200780988c45d910353bd75ab1744f22b768af813cd3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"18ad61c03544d28ce39b200780988c45d910353bd75ab1744f22b768af813cd3","first_computed_at":"2026-07-05T06:05:27.571080Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:05:27.571080Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NLTLNQdWyF6LFQ4ji2qoyDgRqh8GTVbw58Py4zpU7tE+FqdYxHM5X/HzskYc7Pf0R9x4QUZmPxW1xERXT50DBg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:05:27.571571Z","signed_message":"canonical_sha256_bytes"},"source_id":"2206.11871","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:79b66dfd46d972cbd1e9e08f383eb0af2330e3de4a981adee0d9501d1a351ee1","sha256:4bbd3b3ec0b1b072e46a42a8a4f1bb01abef195876baa64070b4f4fa0d69a7ff"],"state_sha256":"b59d0f77a9d16fc834e78940458747b98472ab04d7e7c2e48b8fe36e92a00fb5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MnavpA2n1UueBlQ46ot6/mLNqyzNOKB0NiH3g8NfWs0fc2Aoo6RtyAT9Nix030GxTO26qJn0BKA8v9MQZhdxCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T12:58:53.300713Z","bundle_sha256":"8cb3027c35d670be6ab13f65f6c3627d9f38256e8c36a723d7805772c6c56422"}}