{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2016:EZT5KX4H23E6SIYHTBLJDBAAVI","short_pith_number":"pith:EZT5KX4H","canonical_record":{"source":{"id":"1611.02796","kind":"arxiv","version":9},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-09T01:46:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"056ae21c44a8f8bf0bee5126878617ccfb343a168f6880258e62c43c173f521c","abstract_canon_sha256":"d4e340dc4e9b97cdfaffca47c670812b32e0d337c190193269f77730880cb832"},"schema_version":"1.0"},"canonical_sha256":"2667d55f87d6c9e923079856918400aa3da7c303cb5c54db3b1e36a1ab5736f0","source":{"kind":"arxiv","id":"1611.02796","version":9},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1611.02796","created_at":"2026-05-18T00:32:44Z"},{"alias_kind":"arxiv_version","alias_value":"1611.02796v9","created_at":"2026-05-18T00:32:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1611.02796","created_at":"2026-05-18T00:32:44Z"},{"alias_kind":"pith_short_12","alias_value":"EZT5KX4H23E6","created_at":"2026-05-18T12:30:15Z"},{"alias_kind":"pith_short_16","alias_value":"EZT5KX4H23E6SIYH","created_at":"2026-05-18T12:30:15Z"},{"alias_kind":"pith_short_8","alias_value":"EZT5KX4H","created_at":"2026-05-18T12:30:15Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2016:EZT5KX4H23E6SIYHTBLJDBAAVI","target":"record","payload":{"canonical_record":{"source":{"id":"1611.02796","kind":"arxiv","version":9},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-09T01:46:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"056ae21c44a8f8bf0bee5126878617ccfb343a168f6880258e62c43c173f521c","abstract_canon_sha256":"d4e340dc4e9b97cdfaffca47c670812b32e0d337c190193269f77730880cb832"},"schema_version":"1.0"},"canonical_sha256":"2667d55f87d6c9e923079856918400aa3da7c303cb5c54db3b1e36a1ab5736f0","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:32:44.214013Z","signature_b64":"lha6Upm47cSWVN9cVLq6TEcArwu2FweI/UUkQeQHu4kWgjuVDnA1v89fzUkxIkFiC536qLi3e2nia9YG8W2KCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2667d55f87d6c9e923079856918400aa3da7c303cb5c54db3b1e36a1ab5736f0","last_reissued_at":"2026-05-18T00:32:44.213393Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:32:44.213393Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1611.02796","source_version":9,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:32:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"B7/557y9a1OoxAJjJh+CpyH6GxIoxD++T3io3FAVLXm9Makaw9IshV254r4VvSTtodPJ0oq4dtlamFDbaJkVCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T14:30:44.777624Z"},"content_sha256":"7a5ac11095b56ddd5a486c4ad405102b638b00ac1e1a2078e5a7bf13ed7f61f0","schema_version":"1.0","event_id":"sha256:7a5ac11095b56ddd5a486c4ad405102b638b00ac1e1a2078e5a7bf13ed7f61f0"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2016:EZT5KX4H23E6SIYHTBLJDBAAVI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sequence Tutor: Conservative Fine-Tuning of Sequence Generation Models with KL-control","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Douglas Eck, Dzmitry Bahdanau, Jos\\'e Miguel Hern\\'andez-Lobato, Natasha Jaques, Richard E. Turner, Shixiang Gu","submitted_at":"2016-11-09T01:46:32Z","abstract_excerpt":"This paper proposes a general method for improving the structure and quality of sequences generated by a recurrent neural network (RNN), while maintaining information originally learned from data, as well as sample diversity. An RNN is first pre-trained on data using maximum likelihood estimation (MLE), and the probability distribution over the next token in the sequence learned by this model is treated as a prior policy. Another RNN is then trained using reinforcement learning (RL) to generate higher-quality outputs that account for domain-specific incentives while retaining proximity to the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1611.02796","kind":"arxiv","version":9},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"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":"2026-05-18T00:32:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lT3317F3tw2VxGVU2FTI67imQJTZvwGuZ0wc5YQcIqd/dkvB3c1VfvsvD2UsDlboCQBwvgWlMIYWb8CiUw3mDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T14:30:44.778174Z"},"content_sha256":"2a5f02b1f9736b99aa0eb98034027516f57956547ffeee139494d8a54777bcfe","schema_version":"1.0","event_id":"sha256:2a5f02b1f9736b99aa0eb98034027516f57956547ffeee139494d8a54777bcfe"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EZT5KX4H23E6SIYHTBLJDBAAVI/bundle.json","state_url":"https://pith.science/pith/EZT5KX4H23E6SIYHTBLJDBAAVI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EZT5KX4H23E6SIYHTBLJDBAAVI/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":"2026-06-07T14:30:44Z","links":{"resolver":"https://pith.science/pith/EZT5KX4H23E6SIYHTBLJDBAAVI","bundle":"https://pith.science/pith/EZT5KX4H23E6SIYHTBLJDBAAVI/bundle.json","state":"https://pith.science/pith/EZT5KX4H23E6SIYHTBLJDBAAVI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EZT5KX4H23E6SIYHTBLJDBAAVI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2016:EZT5KX4H23E6SIYHTBLJDBAAVI","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":"d4e340dc4e9b97cdfaffca47c670812b32e0d337c190193269f77730880cb832","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-09T01:46:32Z","title_canon_sha256":"056ae21c44a8f8bf0bee5126878617ccfb343a168f6880258e62c43c173f521c"},"schema_version":"1.0","source":{"id":"1611.02796","kind":"arxiv","version":9}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1611.02796","created_at":"2026-05-18T00:32:44Z"},{"alias_kind":"arxiv_version","alias_value":"1611.02796v9","created_at":"2026-05-18T00:32:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1611.02796","created_at":"2026-05-18T00:32:44Z"},{"alias_kind":"pith_short_12","alias_value":"EZT5KX4H23E6","created_at":"2026-05-18T12:30:15Z"},{"alias_kind":"pith_short_16","alias_value":"EZT5KX4H23E6SIYH","created_at":"2026-05-18T12:30:15Z"},{"alias_kind":"pith_short_8","alias_value":"EZT5KX4H","created_at":"2026-05-18T12:30:15Z"}],"graph_snapshots":[{"event_id":"sha256:2a5f02b1f9736b99aa0eb98034027516f57956547ffeee139494d8a54777bcfe","target":"graph","created_at":"2026-05-18T00:32:44Z","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"},"paper":{"abstract_excerpt":"This paper proposes a general method for improving the structure and quality of sequences generated by a recurrent neural network (RNN), while maintaining information originally learned from data, as well as sample diversity. An RNN is first pre-trained on data using maximum likelihood estimation (MLE), and the probability distribution over the next token in the sequence learned by this model is treated as a prior policy. Another RNN is then trained using reinforcement learning (RL) to generate higher-quality outputs that account for domain-specific incentives while retaining proximity to the ","authors_text":"Douglas Eck, Dzmitry Bahdanau, Jos\\'e Miguel Hern\\'andez-Lobato, Natasha Jaques, Richard E. Turner, Shixiang Gu","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-09T01:46:32Z","title":"Sequence Tutor: Conservative Fine-Tuning of Sequence Generation Models with KL-control"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1611.02796","kind":"arxiv","version":9},"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":"sha256:7a5ac11095b56ddd5a486c4ad405102b638b00ac1e1a2078e5a7bf13ed7f61f0","target":"record","created_at":"2026-05-18T00:32:44Z","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":"d4e340dc4e9b97cdfaffca47c670812b32e0d337c190193269f77730880cb832","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2016-11-09T01:46:32Z","title_canon_sha256":"056ae21c44a8f8bf0bee5126878617ccfb343a168f6880258e62c43c173f521c"},"schema_version":"1.0","source":{"id":"1611.02796","kind":"arxiv","version":9}},"canonical_sha256":"2667d55f87d6c9e923079856918400aa3da7c303cb5c54db3b1e36a1ab5736f0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2667d55f87d6c9e923079856918400aa3da7c303cb5c54db3b1e36a1ab5736f0","first_computed_at":"2026-05-18T00:32:44.213393Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:32:44.213393Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lha6Upm47cSWVN9cVLq6TEcArwu2FweI/UUkQeQHu4kWgjuVDnA1v89fzUkxIkFiC536qLi3e2nia9YG8W2KCQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:32:44.214013Z","signed_message":"canonical_sha256_bytes"},"source_id":"1611.02796","source_kind":"arxiv","source_version":9}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7a5ac11095b56ddd5a486c4ad405102b638b00ac1e1a2078e5a7bf13ed7f61f0","sha256:2a5f02b1f9736b99aa0eb98034027516f57956547ffeee139494d8a54777bcfe"],"state_sha256":"8cebd37b1f7c201fdedef6174158266cc83732b3eba80b8cf81e5ce914e26976"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sviqvGFem4PpmCnrSBrCZPrm886bKTAwLEOqDC9XPSTmqbNoCz79nsLT2BMghJKcquQ2NJE+XFoEBcOjk/xtDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-07T14:30:44.781324Z","bundle_sha256":"16391b4b09666a94a41dc15b0dffbcb63db0b67bd27aadaab737a974ed6b5dfa"}}