{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:7U3DWO3MMBQLN6KIZYLV3SYJF6","short_pith_number":"pith:7U3DWO3M","canonical_record":{"source":{"id":"2007.06018","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-07-12T15:31:24Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"f37b33860a0d77b12398d3279e60dd5a0b8a4ec939cf539359bbe316c1139fec","abstract_canon_sha256":"6e9ae7c3666fa71c5d8ab4484d3e2e47375464935244b06a5e62b6ea366f8cc6"},"schema_version":"1.0"},"canonical_sha256":"fd363b3b6c6060b6f948ce175dcb092f953e6e7a84fd50c41581d2f61ad59356","source":{"kind":"arxiv","id":"2007.06018","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2007.06018","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"arxiv_version","alias_value":"2007.06018v1","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.06018","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"pith_short_12","alias_value":"7U3DWO3MMBQL","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"pith_short_16","alias_value":"7U3DWO3MMBQLN6KI","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"pith_short_8","alias_value":"7U3DWO3M","created_at":"2026-07-05T01:18:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:7U3DWO3MMBQLN6KIZYLV3SYJF6","target":"record","payload":{"canonical_record":{"source":{"id":"2007.06018","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-07-12T15:31:24Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"f37b33860a0d77b12398d3279e60dd5a0b8a4ec939cf539359bbe316c1139fec","abstract_canon_sha256":"6e9ae7c3666fa71c5d8ab4484d3e2e47375464935244b06a5e62b6ea366f8cc6"},"schema_version":"1.0"},"canonical_sha256":"fd363b3b6c6060b6f948ce175dcb092f953e6e7a84fd50c41581d2f61ad59356","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:18:09.163627Z","signature_b64":"3i3N8tbMNNYFlLbIp1H9/qgweJUAYcl2kYnHJKTl3EP3302xRvnDvDcYoZnElb1gBtVpDbn6gpoxzdwxLjg/DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fd363b3b6c6060b6f948ce175dcb092f953e6e7a84fd50c41581d2f61ad59356","last_reissued_at":"2026-07-05T01:18:09.163280Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:18:09.163280Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2007.06018","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":"2026-07-05T01:18:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3a2paskrQHrPsfcnNvsKxGgCikmUUsU1EvLmBQDw/C2Ut1gwV1320sL8Y+3TOOCI+/RRtLGmO4FDrrE0MLLdAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T06:34:09.897429Z"},"content_sha256":"d544bffb710e7a426cf34ec5dd8d0e081831a7b8917af739ebcd0ffbdf8e2571","schema_version":"1.0","event_id":"sha256:d544bffb710e7a426cf34ec5dd8d0e081831a7b8917af739ebcd0ffbdf8e2571"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:7U3DWO3MMBQLN6KIZYLV3SYJF6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Improving Maximum Likelihood Training for Text Generation with Density Ratio Estimation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"stat.ML","authors_text":"Hao Zhou, Lantao Yu, Lei Li, Mingxuan Wang, Ning Miao, Yuxuan Song","submitted_at":"2020-07-12T15:31:24Z","abstract_excerpt":"Auto-regressive sequence generative models trained by Maximum Likelihood Estimation suffer the exposure bias problem in practical finite sample scenarios. The crux is that the number of training samples for Maximum Likelihood Estimation is usually limited and the input data distributions are different at training and inference stages. Many method shave been proposed to solve the above problem (Yu et al., 2017; Lu et al., 2018), which relies on sampling from the non-stationary model distribution and suffers from high variance or biased estimations. In this paper, we propose{\\psi}-MLE, a new tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.06018","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/2007.06018/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":"2026-07-05T01:18:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JtyUIY+/DDScQ+TlZWCW41NSpfIb967WSQByi3OVJtCet6GKAIGC6vVdF1jbqkl8i/qOP+LDsk/1mG+rGS5VAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T06:34:09.898312Z"},"content_sha256":"8ec87e204115e967c0a3517ff975ba30945f5ef7aa053ccfe9fa8598c388a346","schema_version":"1.0","event_id":"sha256:8ec87e204115e967c0a3517ff975ba30945f5ef7aa053ccfe9fa8598c388a346"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7U3DWO3MMBQLN6KIZYLV3SYJF6/bundle.json","state_url":"https://pith.science/pith/7U3DWO3MMBQLN6KIZYLV3SYJF6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7U3DWO3MMBQLN6KIZYLV3SYJF6/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":"2026-08-01T06:34:09Z","links":{"resolver":"https://pith.science/pith/7U3DWO3MMBQLN6KIZYLV3SYJF6","bundle":"https://pith.science/pith/7U3DWO3MMBQLN6KIZYLV3SYJF6/bundle.json","state":"https://pith.science/pith/7U3DWO3MMBQLN6KIZYLV3SYJF6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7U3DWO3MMBQLN6KIZYLV3SYJF6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:7U3DWO3MMBQLN6KIZYLV3SYJF6","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":"6e9ae7c3666fa71c5d8ab4484d3e2e47375464935244b06a5e62b6ea366f8cc6","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-07-12T15:31:24Z","title_canon_sha256":"f37b33860a0d77b12398d3279e60dd5a0b8a4ec939cf539359bbe316c1139fec"},"schema_version":"1.0","source":{"id":"2007.06018","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2007.06018","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"arxiv_version","alias_value":"2007.06018v1","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2007.06018","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"pith_short_12","alias_value":"7U3DWO3MMBQL","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"pith_short_16","alias_value":"7U3DWO3MMBQLN6KI","created_at":"2026-07-05T01:18:09Z"},{"alias_kind":"pith_short_8","alias_value":"7U3DWO3M","created_at":"2026-07-05T01:18:09Z"}],"graph_snapshots":[{"event_id":"sha256:8ec87e204115e967c0a3517ff975ba30945f5ef7aa053ccfe9fa8598c388a346","target":"graph","created_at":"2026-07-05T01:18:09Z","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/2007.06018/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Auto-regressive sequence generative models trained by Maximum Likelihood Estimation suffer the exposure bias problem in practical finite sample scenarios. The crux is that the number of training samples for Maximum Likelihood Estimation is usually limited and the input data distributions are different at training and inference stages. Many method shave been proposed to solve the above problem (Yu et al., 2017; Lu et al., 2018), which relies on sampling from the non-stationary model distribution and suffers from high variance or biased estimations. In this paper, we propose{\\psi}-MLE, a new tra","authors_text":"Hao Zhou, Lantao Yu, Lei Li, Mingxuan Wang, Ning Miao, Yuxuan Song","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-07-12T15:31:24Z","title":"Improving Maximum Likelihood Training for Text Generation with Density Ratio Estimation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2007.06018","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:d544bffb710e7a426cf34ec5dd8d0e081831a7b8917af739ebcd0ffbdf8e2571","target":"record","created_at":"2026-07-05T01:18:09Z","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":"6e9ae7c3666fa71c5d8ab4484d3e2e47375464935244b06a5e62b6ea366f8cc6","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2020-07-12T15:31:24Z","title_canon_sha256":"f37b33860a0d77b12398d3279e60dd5a0b8a4ec939cf539359bbe316c1139fec"},"schema_version":"1.0","source":{"id":"2007.06018","kind":"arxiv","version":1}},"canonical_sha256":"fd363b3b6c6060b6f948ce175dcb092f953e6e7a84fd50c41581d2f61ad59356","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fd363b3b6c6060b6f948ce175dcb092f953e6e7a84fd50c41581d2f61ad59356","first_computed_at":"2026-07-05T01:18:09.163280Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T01:18:09.163280Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3i3N8tbMNNYFlLbIp1H9/qgweJUAYcl2kYnHJKTl3EP3302xRvnDvDcYoZnElb1gBtVpDbn6gpoxzdwxLjg/DA==","signature_status":"signed_v1","signed_at":"2026-07-05T01:18:09.163627Z","signed_message":"canonical_sha256_bytes"},"source_id":"2007.06018","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d544bffb710e7a426cf34ec5dd8d0e081831a7b8917af739ebcd0ffbdf8e2571","sha256:8ec87e204115e967c0a3517ff975ba30945f5ef7aa053ccfe9fa8598c388a346"],"state_sha256":"708553daa49406b72f8359901b3259217442dca1dd4fd7c27587bdaa7ac3deae"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"a+eLtX6Q99p5dv4PBGN9YxJAhjvL0j4BVTQf1G8hVAwxAqSURzkkFqOsx/+fdUkgLBeNQ1yfE4yp7+oR4eDvDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T06:34:09.908107Z","bundle_sha256":"73c99de69ff71f44f10b4f44f109a15504b167d7b4f54e7e68b9f4f3a65eada1"}}