{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:34ZZCEDEQHX62Z4BAKTEDX2RQ6","short_pith_number":"pith:34ZZCEDE","canonical_record":{"source":{"id":"1907.12009","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2019-07-28T03:57:41Z","cross_cats_sorted":[],"title_canon_sha256":"137f5b4f2f60207fb9ebc7bae0d19c2c6aedd770877a3bb43dd51711d83e3f1a","abstract_canon_sha256":"eaeeb577a24af1faac0edaf38482275465cfe4d2795c0992c0f6f516aef38b1b"},"schema_version":"1.0"},"canonical_sha256":"df3391106481efed678102a641df518785a8cdce7bb63cd9caf78df7400de581","source":{"kind":"arxiv","id":"1907.12009","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.12009","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"arxiv_version","alias_value":"1907.12009v1","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.12009","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"pith_short_12","alias_value":"34ZZCEDEQHX6","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"pith_short_16","alias_value":"34ZZCEDEQHX62Z4B","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"pith_short_8","alias_value":"34ZZCEDE","created_at":"2026-07-04T23:50:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:34ZZCEDEQHX62Z4BAKTEDX2RQ6","target":"record","payload":{"canonical_record":{"source":{"id":"1907.12009","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2019-07-28T03:57:41Z","cross_cats_sorted":[],"title_canon_sha256":"137f5b4f2f60207fb9ebc7bae0d19c2c6aedd770877a3bb43dd51711d83e3f1a","abstract_canon_sha256":"eaeeb577a24af1faac0edaf38482275465cfe4d2795c0992c0f6f516aef38b1b"},"schema_version":"1.0"},"canonical_sha256":"df3391106481efed678102a641df518785a8cdce7bb63cd9caf78df7400de581","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:50:04.589294Z","signature_b64":"aNJNENc4wDv313AgtafsxCbHKlk45r/S0yVMNyhh+Pd2ja1mdSjfWnWiqEYdEAl5wePpS/I0Dc7aOmtbOi7kDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df3391106481efed678102a641df518785a8cdce7bb63cd9caf78df7400de581","last_reissued_at":"2026-07-04T23:50:04.588819Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:50:04.588819Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1907.12009","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":"2026-07-04T23:50:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NZ5LvQJ9zo+uQFePfyfmg+C0tZ0qTJFxlWE34nUePanfooVsIvutAEydgHEwSm6EFp6hWgzTvYbG1vb1UC/JDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T08:11:18.061709Z"},"content_sha256":"c8c9e4eed1d20f9398843964f70996120cdf915cc31841527a36b859f4082083","schema_version":"1.0","event_id":"sha256:c8c9e4eed1d20f9398843964f70996120cdf915cc31841527a36b859f4082083"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:34ZZCEDEQHX62Z4BAKTEDX2RQ6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Representation Degeneration Problem in Training Natural Language Generation Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Di He, Jun Gao, Liwei Wang, Tao Qin, Tie-Yan Liu, Xu Tan","submitted_at":"2019-07-28T03:57:41Z","abstract_excerpt":"We study an interesting problem in training neural network-based models for natural language generation tasks, which we call the \\emph{representation degeneration problem}. We observe that when training a model for natural language generation tasks through likelihood maximization with the weight tying trick, especially with big training datasets, most of the learnt word embeddings tend to degenerate and be distributed into a narrow cone, which largely limits the representation power of word embeddings. We analyze the conditions and causes of this problem and propose a novel regularization meth"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.12009","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/1907.12009/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":"2026-07-04T23:50:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0l9z/kChHojELnvpMHhhw/dyQNu724ujzNxFdyxS1gv5kUjTSjobJnRPWQqv2zVQzNBELmWpuaSzucyQoeazCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T08:11:18.062223Z"},"content_sha256":"3a103347667730832e8da5cee7c2624a5c4e886c884f83df66b08054c413bce0","schema_version":"1.0","event_id":"sha256:3a103347667730832e8da5cee7c2624a5c4e886c884f83df66b08054c413bce0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/34ZZCEDEQHX62Z4BAKTEDX2RQ6/bundle.json","state_url":"https://pith.science/pith/34ZZCEDEQHX62Z4BAKTEDX2RQ6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/34ZZCEDEQHX62Z4BAKTEDX2RQ6/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":"2026-08-06T08:11:18Z","links":{"resolver":"https://pith.science/pith/34ZZCEDEQHX62Z4BAKTEDX2RQ6","bundle":"https://pith.science/pith/34ZZCEDEQHX62Z4BAKTEDX2RQ6/bundle.json","state":"https://pith.science/pith/34ZZCEDEQHX62Z4BAKTEDX2RQ6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/34ZZCEDEQHX62Z4BAKTEDX2RQ6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:34ZZCEDEQHX62Z4BAKTEDX2RQ6","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":"eaeeb577a24af1faac0edaf38482275465cfe4d2795c0992c0f6f516aef38b1b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2019-07-28T03:57:41Z","title_canon_sha256":"137f5b4f2f60207fb9ebc7bae0d19c2c6aedd770877a3bb43dd51711d83e3f1a"},"schema_version":"1.0","source":{"id":"1907.12009","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1907.12009","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"arxiv_version","alias_value":"1907.12009v1","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1907.12009","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"pith_short_12","alias_value":"34ZZCEDEQHX6","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"pith_short_16","alias_value":"34ZZCEDEQHX62Z4B","created_at":"2026-07-04T23:50:04Z"},{"alias_kind":"pith_short_8","alias_value":"34ZZCEDE","created_at":"2026-07-04T23:50:04Z"}],"graph_snapshots":[{"event_id":"sha256:3a103347667730832e8da5cee7c2624a5c4e886c884f83df66b08054c413bce0","target":"graph","created_at":"2026-07-04T23:50:04Z","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/1907.12009/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We study an interesting problem in training neural network-based models for natural language generation tasks, which we call the \\emph{representation degeneration problem}. We observe that when training a model for natural language generation tasks through likelihood maximization with the weight tying trick, especially with big training datasets, most of the learnt word embeddings tend to degenerate and be distributed into a narrow cone, which largely limits the representation power of word embeddings. We analyze the conditions and causes of this problem and propose a novel regularization meth","authors_text":"Di He, Jun Gao, Liwei Wang, Tao Qin, Tie-Yan Liu, Xu Tan","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2019-07-28T03:57:41Z","title":"Representation Degeneration Problem in Training Natural Language Generation Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1907.12009","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:c8c9e4eed1d20f9398843964f70996120cdf915cc31841527a36b859f4082083","target":"record","created_at":"2026-07-04T23:50:04Z","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":"eaeeb577a24af1faac0edaf38482275465cfe4d2795c0992c0f6f516aef38b1b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2019-07-28T03:57:41Z","title_canon_sha256":"137f5b4f2f60207fb9ebc7bae0d19c2c6aedd770877a3bb43dd51711d83e3f1a"},"schema_version":"1.0","source":{"id":"1907.12009","kind":"arxiv","version":1}},"canonical_sha256":"df3391106481efed678102a641df518785a8cdce7bb63cd9caf78df7400de581","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"df3391106481efed678102a641df518785a8cdce7bb63cd9caf78df7400de581","first_computed_at":"2026-07-04T23:50:04.588819Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-04T23:50:04.588819Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"aNJNENc4wDv313AgtafsxCbHKlk45r/S0yVMNyhh+Pd2ja1mdSjfWnWiqEYdEAl5wePpS/I0Dc7aOmtbOi7kDw==","signature_status":"signed_v1","signed_at":"2026-07-04T23:50:04.589294Z","signed_message":"canonical_sha256_bytes"},"source_id":"1907.12009","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c8c9e4eed1d20f9398843964f70996120cdf915cc31841527a36b859f4082083","sha256:3a103347667730832e8da5cee7c2624a5c4e886c884f83df66b08054c413bce0"],"state_sha256":"a3db65fe89a1abdb21c6a0a3554777585e1e6badca9a95fc7e8e8d6965ff67f8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ocJR/Lxxe8CmK7MGs7yixTezcDDNurWrgnX6VprBVK67JilXvODFAQKMsiGsnz9EEa+yBBZNzQUaTHY4VVleDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T08:11:18.066507Z","bundle_sha256":"9f2a5bc08e080d8353e650d3a357528118a7daef4f2f58ed156fde1cdd505730"}}