{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:EAIFARRAS7FEXF6J7A66SKKSIH","short_pith_number":"pith:EAIFARRA","canonical_record":{"source":{"id":"1908.04319","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-08-12T18:09:04Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"ca328ae9b9ea70c9060ef8444192bf495bc752e29b0aee28d48e47fff71f8502","abstract_canon_sha256":"5b28c5e665cba7c5a0641931007cb50859013c4b1ae741cb82067726d95a0518"},"schema_version":"1.0"},"canonical_sha256":"201050462097ca4b97c9f83de9295241c120c8cdd9d1c4c234b9999121d9a252","source":{"kind":"arxiv","id":"1908.04319","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1908.04319","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"arxiv_version","alias_value":"1908.04319v2","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.04319","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"pith_short_12","alias_value":"EAIFARRAS7FE","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"pith_short_16","alias_value":"EAIFARRAS7FEXF6J","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"pith_short_8","alias_value":"EAIFARRA","created_at":"2026-07-05T00:07:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:EAIFARRAS7FEXF6J7A66SKKSIH","target":"record","payload":{"canonical_record":{"source":{"id":"1908.04319","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-08-12T18:09:04Z","cross_cats_sorted":["cs.CL","stat.ML"],"title_canon_sha256":"ca328ae9b9ea70c9060ef8444192bf495bc752e29b0aee28d48e47fff71f8502","abstract_canon_sha256":"5b28c5e665cba7c5a0641931007cb50859013c4b1ae741cb82067726d95a0518"},"schema_version":"1.0"},"canonical_sha256":"201050462097ca4b97c9f83de9295241c120c8cdd9d1c4c234b9999121d9a252","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:07:37.143529Z","signature_b64":"djFY9lq8/N+Nb41aa5hPEkqfUAfHdTP0GNj9nP78V5WdsaMdy5DmJB87W9oGZsrF8/oVW9NhcwHJcT61hlCrAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"201050462097ca4b97c9f83de9295241c120c8cdd9d1c4c234b9999121d9a252","last_reissued_at":"2026-07-05T00:07:37.143055Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:07:37.143055Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1908.04319","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":"2026-07-05T00:07:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"WLZzg7iEDCouR8PtP3P5Y8+8nkRhCKX/xP4WzidAl3fJulBtD+6I01svOij1ATlG66zCc80uS+woxhu6DfPICA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-13T19:06:36.813108Z"},"content_sha256":"6d94384d99fea78f5447d457dcb899faa9c64ed08f7d78d594b65506a88c5b83","schema_version":"1.0","event_id":"sha256:6d94384d99fea78f5447d457dcb899faa9c64ed08f7d78d594b65506a88c5b83"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:EAIFARRAS7FEXF6J7A66SKKSIH","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Neural Text Generation with Unlikelihood Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","stat.ML"],"primary_cat":"cs.LG","authors_text":"Emily Dinan, Ilia Kulikov, Jason Weston, Kyunghyun Cho, Sean Welleck, Stephen Roller","submitted_at":"2019-08-12T18:09:04Z","abstract_excerpt":"Neural text generation is a key tool in natural language applications, but it is well known there are major problems at its core. In particular, standard likelihood training and decoding leads to dull and repetitive outputs. While some post-hoc fixes have been proposed, in particular top-$k$ and nucleus sampling, they do not address the fact that the token-level probabilities predicted by the model are poor. In this paper we show that the likelihood objective itself is at fault, resulting in a model that assigns too much probability to sequences containing repeats and frequent words, unlike th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.04319","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/1908.04319/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":"2026-07-05T00:07:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J6IJ6CGJaRH0mfjS6S8LzP257UrCJ8QIa+v+YGpCERfVEiJDBoSr10kU3o9SUakEqzdJOrmG9wMwcIY7BlT8Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-13T19:06:36.813500Z"},"content_sha256":"ce0052344f78351eaf6e77a5ca8b20a2e7c9c0380682fe204d71052bd67d991e","schema_version":"1.0","event_id":"sha256:ce0052344f78351eaf6e77a5ca8b20a2e7c9c0380682fe204d71052bd67d991e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EAIFARRAS7FEXF6J7A66SKKSIH/bundle.json","state_url":"https://pith.science/pith/EAIFARRAS7FEXF6J7A66SKKSIH/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EAIFARRAS7FEXF6J7A66SKKSIH/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":"2026-07-13T19:06:36Z","links":{"resolver":"https://pith.science/pith/EAIFARRAS7FEXF6J7A66SKKSIH","bundle":"https://pith.science/pith/EAIFARRAS7FEXF6J7A66SKKSIH/bundle.json","state":"https://pith.science/pith/EAIFARRAS7FEXF6J7A66SKKSIH/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EAIFARRAS7FEXF6J7A66SKKSIH/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:EAIFARRAS7FEXF6J7A66SKKSIH","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":"5b28c5e665cba7c5a0641931007cb50859013c4b1ae741cb82067726d95a0518","cross_cats_sorted":["cs.CL","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-08-12T18:09:04Z","title_canon_sha256":"ca328ae9b9ea70c9060ef8444192bf495bc752e29b0aee28d48e47fff71f8502"},"schema_version":"1.0","source":{"id":"1908.04319","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1908.04319","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"arxiv_version","alias_value":"1908.04319v2","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.04319","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"pith_short_12","alias_value":"EAIFARRAS7FE","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"pith_short_16","alias_value":"EAIFARRAS7FEXF6J","created_at":"2026-07-05T00:07:37Z"},{"alias_kind":"pith_short_8","alias_value":"EAIFARRA","created_at":"2026-07-05T00:07:37Z"}],"graph_snapshots":[{"event_id":"sha256:ce0052344f78351eaf6e77a5ca8b20a2e7c9c0380682fe204d71052bd67d991e","target":"graph","created_at":"2026-07-05T00:07:37Z","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/1908.04319/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Neural text generation is a key tool in natural language applications, but it is well known there are major problems at its core. In particular, standard likelihood training and decoding leads to dull and repetitive outputs. While some post-hoc fixes have been proposed, in particular top-$k$ and nucleus sampling, they do not address the fact that the token-level probabilities predicted by the model are poor. In this paper we show that the likelihood objective itself is at fault, resulting in a model that assigns too much probability to sequences containing repeats and frequent words, unlike th","authors_text":"Emily Dinan, Ilia Kulikov, Jason Weston, Kyunghyun Cho, Sean Welleck, Stephen Roller","cross_cats":["cs.CL","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-08-12T18:09:04Z","title":"Neural Text Generation with Unlikelihood Training"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.04319","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":"sha256:6d94384d99fea78f5447d457dcb899faa9c64ed08f7d78d594b65506a88c5b83","target":"record","created_at":"2026-07-05T00:07:37Z","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":"5b28c5e665cba7c5a0641931007cb50859013c4b1ae741cb82067726d95a0518","cross_cats_sorted":["cs.CL","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-08-12T18:09:04Z","title_canon_sha256":"ca328ae9b9ea70c9060ef8444192bf495bc752e29b0aee28d48e47fff71f8502"},"schema_version":"1.0","source":{"id":"1908.04319","kind":"arxiv","version":2}},"canonical_sha256":"201050462097ca4b97c9f83de9295241c120c8cdd9d1c4c234b9999121d9a252","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"201050462097ca4b97c9f83de9295241c120c8cdd9d1c4c234b9999121d9a252","first_computed_at":"2026-07-05T00:07:37.143055Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:07:37.143055Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"djFY9lq8/N+Nb41aa5hPEkqfUAfHdTP0GNj9nP78V5WdsaMdy5DmJB87W9oGZsrF8/oVW9NhcwHJcT61hlCrAg==","signature_status":"signed_v1","signed_at":"2026-07-05T00:07:37.143529Z","signed_message":"canonical_sha256_bytes"},"source_id":"1908.04319","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6d94384d99fea78f5447d457dcb899faa9c64ed08f7d78d594b65506a88c5b83","sha256:ce0052344f78351eaf6e77a5ca8b20a2e7c9c0380682fe204d71052bd67d991e"],"state_sha256":"c7e24895d1365d668a8256f1da8acd07372b8d7d2536529dc8f6e67a579211ea"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uwzhU1W5X+uGjgLdptCLH87ntEBcIeJQbp4GpSgBIvoJ7o2T000KQ2KBeivY/ft+TKFY6cjANFhDqPU3Y3MEBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-13T19:06:36.815525Z","bundle_sha256":"7bb457b677bbebf6fc4b7f1214e5f234427fb5f658c43eeaaf4b9d805dba0e60"}}