{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:4TUPT7L2UHSC7OBKE3IVK4K666","short_pith_number":"pith:4TUPT7L2","canonical_record":{"source":{"id":"1808.07982","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-24T02:14:43Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"61224080f1a1ae947cbab93c00a7d29a1eeb081a7fd0c137278c6a65f9250e19","abstract_canon_sha256":"249d361ef9d861ce4282e24080b48cbd4c9d303cd5ebe881c142186d98cba9fe"},"schema_version":"1.0"},"canonical_sha256":"e4e8f9fd7aa1e42fb82a26d155715ef7a2aa09cb810bfc147c4f8abb9e920247","source":{"kind":"arxiv","id":"1808.07982","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1808.07982","created_at":"2026-05-18T00:07:22Z"},{"alias_kind":"arxiv_version","alias_value":"1808.07982v1","created_at":"2026-05-18T00:07:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.07982","created_at":"2026-05-18T00:07:22Z"},{"alias_kind":"pith_short_12","alias_value":"4TUPT7L2UHSC","created_at":"2026-05-18T12:32:05Z"},{"alias_kind":"pith_short_16","alias_value":"4TUPT7L2UHSC7OBK","created_at":"2026-05-18T12:32:05Z"},{"alias_kind":"pith_short_8","alias_value":"4TUPT7L2","created_at":"2026-05-18T12:32:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:4TUPT7L2UHSC7OBKE3IVK4K666","target":"record","payload":{"canonical_record":{"source":{"id":"1808.07982","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-24T02:14:43Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"61224080f1a1ae947cbab93c00a7d29a1eeb081a7fd0c137278c6a65f9250e19","abstract_canon_sha256":"249d361ef9d861ce4282e24080b48cbd4c9d303cd5ebe881c142186d98cba9fe"},"schema_version":"1.0"},"canonical_sha256":"e4e8f9fd7aa1e42fb82a26d155715ef7a2aa09cb810bfc147c4f8abb9e920247","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:07:22.870860Z","signature_b64":"BbtJ3+avqsHijh0ujOvePbWvL7VPMhpZ6Z+rsFlhSo2C21OUicsesDLuoFTfnIE6f9uBpJr4h892INQYrF5AAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4e8f9fd7aa1e42fb82a26d155715ef7a2aa09cb810bfc147c4f8abb9e920247","last_reissued_at":"2026-05-18T00:07:22.870349Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:07:22.870349Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1808.07982","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":"2026-05-18T00:07:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"D2SK02x48wAy2q0WYhBQAsYwJmODjObUPqY9Jb3N3asfR7nOUYXUmFrfUIFwVOKcjetSoLKPkQauFi2Kq07xAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-12T06:48:20.307008Z"},"content_sha256":"f08596e46f6a76f79d88fe7f3104f13bf23f7768ec630b4487f0bbaf5d8ff556","schema_version":"1.0","event_id":"sha256:f08596e46f6a76f79d88fe7f3104f13bf23f7768ec630b4487f0bbaf5d8ff556"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:4TUPT7L2UHSC7OBKE3IVK4K666","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Proximal Policy Optimization and its Dynamic Version for Sequence Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Hung-yi Lee, Jinzhi Zhang, Yi-Lin Tuan, Yujia Li","submitted_at":"2018-08-24T02:14:43Z","abstract_excerpt":"In sequence generation task, many works use policy gradient for model optimization to tackle the intractable backpropagation issue when maximizing the non-differentiable evaluation metrics or fooling the discriminator in adversarial learning. In this paper, we replace policy gradient with proximal policy optimization (PPO), which is a proved more efficient reinforcement learning algorithm, and propose a dynamic approach for PPO (PPO-dynamic). We demonstrate the efficacy of PPO and PPO-dynamic on conditional sequence generation tasks including synthetic experiment and chit-chat chatbot. The res"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.07982","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":""},"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":"2026-05-18T00:07:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jMu2E7yLljNdnGCSoiBemaSHyIBm5vByGqulGbPKnwV1EODw4LfmczhxJvFLIDLMT6IBqwUTEkCLUg9MGXrvBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-12T06:48:20.307663Z"},"content_sha256":"890fa3147109c3a5fde6581786a4545fe19fe81ca344a9e85089d90d434c6dbd","schema_version":"1.0","event_id":"sha256:890fa3147109c3a5fde6581786a4545fe19fe81ca344a9e85089d90d434c6dbd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4TUPT7L2UHSC7OBKE3IVK4K666/bundle.json","state_url":"https://pith.science/pith/4TUPT7L2UHSC7OBKE3IVK4K666/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4TUPT7L2UHSC7OBKE3IVK4K666/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":"2026-06-12T06:48:20Z","links":{"resolver":"https://pith.science/pith/4TUPT7L2UHSC7OBKE3IVK4K666","bundle":"https://pith.science/pith/4TUPT7L2UHSC7OBKE3IVK4K666/bundle.json","state":"https://pith.science/pith/4TUPT7L2UHSC7OBKE3IVK4K666/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4TUPT7L2UHSC7OBKE3IVK4K666/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:4TUPT7L2UHSC7OBKE3IVK4K666","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":"249d361ef9d861ce4282e24080b48cbd4c9d303cd5ebe881c142186d98cba9fe","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-24T02:14:43Z","title_canon_sha256":"61224080f1a1ae947cbab93c00a7d29a1eeb081a7fd0c137278c6a65f9250e19"},"schema_version":"1.0","source":{"id":"1808.07982","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1808.07982","created_at":"2026-05-18T00:07:22Z"},{"alias_kind":"arxiv_version","alias_value":"1808.07982v1","created_at":"2026-05-18T00:07:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.07982","created_at":"2026-05-18T00:07:22Z"},{"alias_kind":"pith_short_12","alias_value":"4TUPT7L2UHSC","created_at":"2026-05-18T12:32:05Z"},{"alias_kind":"pith_short_16","alias_value":"4TUPT7L2UHSC7OBK","created_at":"2026-05-18T12:32:05Z"},{"alias_kind":"pith_short_8","alias_value":"4TUPT7L2","created_at":"2026-05-18T12:32:05Z"}],"graph_snapshots":[{"event_id":"sha256:890fa3147109c3a5fde6581786a4545fe19fe81ca344a9e85089d90d434c6dbd","target":"graph","created_at":"2026-05-18T00:07:22Z","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":"In sequence generation task, many works use policy gradient for model optimization to tackle the intractable backpropagation issue when maximizing the non-differentiable evaluation metrics or fooling the discriminator in adversarial learning. In this paper, we replace policy gradient with proximal policy optimization (PPO), which is a proved more efficient reinforcement learning algorithm, and propose a dynamic approach for PPO (PPO-dynamic). We demonstrate the efficacy of PPO and PPO-dynamic on conditional sequence generation tasks including synthetic experiment and chit-chat chatbot. The res","authors_text":"Hung-yi Lee, Jinzhi Zhang, Yi-Lin Tuan, Yujia Li","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-24T02:14:43Z","title":"Proximal Policy Optimization and its Dynamic Version for Sequence Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.07982","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:f08596e46f6a76f79d88fe7f3104f13bf23f7768ec630b4487f0bbaf5d8ff556","target":"record","created_at":"2026-05-18T00:07:22Z","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":"249d361ef9d861ce4282e24080b48cbd4c9d303cd5ebe881c142186d98cba9fe","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-24T02:14:43Z","title_canon_sha256":"61224080f1a1ae947cbab93c00a7d29a1eeb081a7fd0c137278c6a65f9250e19"},"schema_version":"1.0","source":{"id":"1808.07982","kind":"arxiv","version":1}},"canonical_sha256":"e4e8f9fd7aa1e42fb82a26d155715ef7a2aa09cb810bfc147c4f8abb9e920247","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e4e8f9fd7aa1e42fb82a26d155715ef7a2aa09cb810bfc147c4f8abb9e920247","first_computed_at":"2026-05-18T00:07:22.870349Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:07:22.870349Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"BbtJ3+avqsHijh0ujOvePbWvL7VPMhpZ6Z+rsFlhSo2C21OUicsesDLuoFTfnIE6f9uBpJr4h892INQYrF5AAg==","signature_status":"signed_v1","signed_at":"2026-05-18T00:07:22.870860Z","signed_message":"canonical_sha256_bytes"},"source_id":"1808.07982","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f08596e46f6a76f79d88fe7f3104f13bf23f7768ec630b4487f0bbaf5d8ff556","sha256:890fa3147109c3a5fde6581786a4545fe19fe81ca344a9e85089d90d434c6dbd"],"state_sha256":"08f3221cb7abf042dc99b85ee744998acfc4d4b3411a1cc4119d7482e9970eba"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3PkZrMpbzjH/eJLzjNvxgzvO4/0AJOPDvz+l4XWANhOWJ8TVCkLKKcQ6Azy4G2sh94BbZDHBBgsSdzUPqs5CCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-12T06:48:20.310810Z","bundle_sha256":"a704130edc4e900a19bc74758aec95833cf2dbe98a190278e9e39865878acfbd"}}