{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:EWFWDWNKVAIWMNIUC2QEUNTLIQ","short_pith_number":"pith:EWFWDWNK","canonical_record":{"source":{"id":"2308.02223","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T09:35:45Z","cross_cats_sorted":[],"title_canon_sha256":"af866bbf2a2a633146ad667bb67c293b3f22731604fbaabd42332b0caac4a746","abstract_canon_sha256":"7cf73833268a601c0c9ed87d423fc0ac14d32e9e50c17b7e3a1cf87578810955"},"schema_version":"1.0"},"canonical_sha256":"258b61d9aaa81166351416a04a366b4427e7d274d326c1af4c6c4ec89e595855","source":{"kind":"arxiv","id":"2308.02223","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.02223","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"arxiv_version","alias_value":"2308.02223v1","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.02223","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"pith_short_12","alias_value":"EWFWDWNKVAIW","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"pith_short_16","alias_value":"EWFWDWNKVAIWMNIU","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"pith_short_8","alias_value":"EWFWDWNK","created_at":"2026-07-05T06:37:39Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:EWFWDWNKVAIWMNIUC2QEUNTLIQ","target":"record","payload":{"canonical_record":{"source":{"id":"2308.02223","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T09:35:45Z","cross_cats_sorted":[],"title_canon_sha256":"af866bbf2a2a633146ad667bb67c293b3f22731604fbaabd42332b0caac4a746","abstract_canon_sha256":"7cf73833268a601c0c9ed87d423fc0ac14d32e9e50c17b7e3a1cf87578810955"},"schema_version":"1.0"},"canonical_sha256":"258b61d9aaa81166351416a04a366b4427e7d274d326c1af4c6c4ec89e595855","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:37:39.487340Z","signature_b64":"3q6tCSCMrswgrK2hue+HYyvtkjNJTBXTNqUT7kybTmTT9a8nj39709aupZgBF765wCOWrCTZ1V8pRjey2YuIBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"258b61d9aaa81166351416a04a366b4427e7d274d326c1af4c6c4ec89e595855","last_reissued_at":"2026-07-05T06:37:39.486850Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:37:39.486850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2308.02223","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":"2026-07-05T06:37:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"o1GunTP/Yewe/0NH4hz5WD15saxegCuGGL4r5IbT6JGqVzvkbgIF1eXvZgc7FO8NkIBoOpi8jnlHJ+VRdaMvBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-17T18:50:38.719410Z"},"content_sha256":"fc038b7a5440378a88796b9e411f2951954bf121ddb1ad7f7ad1ebefd33ad0b5","schema_version":"1.0","event_id":"sha256:fc038b7a5440378a88796b9e411f2951954bf121ddb1ad7f7ad1ebefd33ad0b5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:EWFWDWNKVAIWMNIUC2QEUNTLIQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ESRL: Efficient Sampling-based Reinforcement Learning for Sequence Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bei Li, Chenglong Wang, Hang Zhou, Jingbo Zhu, Tongran Liu, Tong Xiao, Yifu Huo, Yimin Hu","submitted_at":"2023-08-04T09:35:45Z","abstract_excerpt":"Applying Reinforcement Learning (RL) to sequence generation models enables the direct optimization of long-term rewards (\\textit{e.g.,} BLEU and human feedback), but typically requires large-scale sampling over a space of action sequences. This is a computational challenge as presented by the practice of sequence generation problems, such as machine translation, where we often deal with a large action space (\\textit{e.g.,} a vocabulary) and a long action sequence (\\textit{e.g.,} a translation). In this work, we introduce two-stage sampling and dynamic sampling approaches to improve the samplin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.02223","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/2308.02223/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":"2026-07-05T06:37:39Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FhALPoN7cQTHFcuvZzFjNork9xUU2R1gFcH+zzP0Xbu5mlmH2PVg+AFe622w3Fm7gx9rbtunyM9cSSPS3mivAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-17T18:50:38.719791Z"},"content_sha256":"0e58e36ca26be11a99e73bd86a25e29d40ac44854459e127819e811e6f790f28","schema_version":"1.0","event_id":"sha256:0e58e36ca26be11a99e73bd86a25e29d40ac44854459e127819e811e6f790f28"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EWFWDWNKVAIWMNIUC2QEUNTLIQ/bundle.json","state_url":"https://pith.science/pith/EWFWDWNKVAIWMNIUC2QEUNTLIQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EWFWDWNKVAIWMNIUC2QEUNTLIQ/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":"2026-07-17T18:50:38Z","links":{"resolver":"https://pith.science/pith/EWFWDWNKVAIWMNIUC2QEUNTLIQ","bundle":"https://pith.science/pith/EWFWDWNKVAIWMNIUC2QEUNTLIQ/bundle.json","state":"https://pith.science/pith/EWFWDWNKVAIWMNIUC2QEUNTLIQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EWFWDWNKVAIWMNIUC2QEUNTLIQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:EWFWDWNKVAIWMNIUC2QEUNTLIQ","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":"7cf73833268a601c0c9ed87d423fc0ac14d32e9e50c17b7e3a1cf87578810955","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T09:35:45Z","title_canon_sha256":"af866bbf2a2a633146ad667bb67c293b3f22731604fbaabd42332b0caac4a746"},"schema_version":"1.0","source":{"id":"2308.02223","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.02223","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"arxiv_version","alias_value":"2308.02223v1","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.02223","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"pith_short_12","alias_value":"EWFWDWNKVAIW","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"pith_short_16","alias_value":"EWFWDWNKVAIWMNIU","created_at":"2026-07-05T06:37:39Z"},{"alias_kind":"pith_short_8","alias_value":"EWFWDWNK","created_at":"2026-07-05T06:37:39Z"}],"graph_snapshots":[{"event_id":"sha256:0e58e36ca26be11a99e73bd86a25e29d40ac44854459e127819e811e6f790f28","target":"graph","created_at":"2026-07-05T06:37:39Z","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/2308.02223/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Applying Reinforcement Learning (RL) to sequence generation models enables the direct optimization of long-term rewards (\\textit{e.g.,} BLEU and human feedback), but typically requires large-scale sampling over a space of action sequences. This is a computational challenge as presented by the practice of sequence generation problems, such as machine translation, where we often deal with a large action space (\\textit{e.g.,} a vocabulary) and a long action sequence (\\textit{e.g.,} a translation). In this work, we introduce two-stage sampling and dynamic sampling approaches to improve the samplin","authors_text":"Bei Li, Chenglong Wang, Hang Zhou, Jingbo Zhu, Tongran Liu, Tong Xiao, Yifu Huo, Yimin Hu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T09:35:45Z","title":"ESRL: Efficient Sampling-based Reinforcement Learning for Sequence Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.02223","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:fc038b7a5440378a88796b9e411f2951954bf121ddb1ad7f7ad1ebefd33ad0b5","target":"record","created_at":"2026-07-05T06:37:39Z","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":"7cf73833268a601c0c9ed87d423fc0ac14d32e9e50c17b7e3a1cf87578810955","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T09:35:45Z","title_canon_sha256":"af866bbf2a2a633146ad667bb67c293b3f22731604fbaabd42332b0caac4a746"},"schema_version":"1.0","source":{"id":"2308.02223","kind":"arxiv","version":1}},"canonical_sha256":"258b61d9aaa81166351416a04a366b4427e7d274d326c1af4c6c4ec89e595855","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"258b61d9aaa81166351416a04a366b4427e7d274d326c1af4c6c4ec89e595855","first_computed_at":"2026-07-05T06:37:39.486850Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:37:39.486850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3q6tCSCMrswgrK2hue+HYyvtkjNJTBXTNqUT7kybTmTT9a8nj39709aupZgBF765wCOWrCTZ1V8pRjey2YuIBg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:37:39.487340Z","signed_message":"canonical_sha256_bytes"},"source_id":"2308.02223","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fc038b7a5440378a88796b9e411f2951954bf121ddb1ad7f7ad1ebefd33ad0b5","sha256:0e58e36ca26be11a99e73bd86a25e29d40ac44854459e127819e811e6f790f28"],"state_sha256":"75e26a3a72065fe1e9773fae976d2beea87c3b3057e65bc6d9d20a9bb653791b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xg14KkHMW+4V4MU3DBGsvhGj+QyX21BcGXR5UzYnKcwwKI0u1LmDSqlgwUnkzx3WJShvp+QIs3pT0tXd60++Aw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-17T18:50:38.722670Z","bundle_sha256":"ea661d63e30c1ad33dddc3a182dd417edfbbc41505132f1c121ee600cb675030"}}