{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:AUUGGRYYUL3DI2A65J5R2Y22TL","short_pith_number":"pith:AUUGGRYY","canonical_record":{"source":{"id":"2201.11627","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-01-26T07:47:27Z","cross_cats_sorted":["cs.AI","cs.SD"],"title_canon_sha256":"82e15d21826ebc9143f3b60b307fb6eaffa02554f72ffa36976a65112668d032","abstract_canon_sha256":"fa8f0783d5c005a47fc4b3a275596c0f244984e09b5679f0ba4c666beb3c8527"},"schema_version":"1.0"},"canonical_sha256":"0528634718a2f634681eea7b1d635a9ad8ee400c152ed5face1d3807ddee763e","source":{"kind":"arxiv","id":"2201.11627","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.11627","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"arxiv_version","alias_value":"2201.11627v3","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.11627","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"pith_short_12","alias_value":"AUUGGRYYUL3D","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"pith_short_16","alias_value":"AUUGGRYYUL3DI2A6","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"pith_short_8","alias_value":"AUUGGRYY","created_at":"2026-07-05T05:12:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:AUUGGRYYUL3DI2A65J5R2Y22TL","target":"record","payload":{"canonical_record":{"source":{"id":"2201.11627","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-01-26T07:47:27Z","cross_cats_sorted":["cs.AI","cs.SD"],"title_canon_sha256":"82e15d21826ebc9143f3b60b307fb6eaffa02554f72ffa36976a65112668d032","abstract_canon_sha256":"fa8f0783d5c005a47fc4b3a275596c0f244984e09b5679f0ba4c666beb3c8527"},"schema_version":"1.0"},"canonical_sha256":"0528634718a2f634681eea7b1d635a9ad8ee400c152ed5face1d3807ddee763e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:27.198609Z","signature_b64":"/YA4pyb3MdUwmiDCNfU+Ek8xPiYHr0PM6PV49WbcwBCZuK0rvoaRsS4qFJ8Ou4uqyPJT2MjoOTcr3TfJL+omAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0528634718a2f634681eea7b1d635a9ad8ee400c152ed5face1d3807ddee763e","last_reissued_at":"2026-07-05T05:12:27.198099Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:27.198099Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2201.11627","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:12:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cC8o3AahfDDUKdIoq04PEAWFw2KaMVktfVeyPdEGdRXsgRedIqNtPa2RXYdPtYbdrU51yjm8a13H4+J9/KR9Dw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-15T23:48:27.444616Z"},"content_sha256":"6b5805a5e35b96dad0b0be757102372e7c7efa924261b96ed4c168d462b87ca2","schema_version":"1.0","event_id":"sha256:6b5805a5e35b96dad0b0be757102372e7c7efa924261b96ed4c168d462b87ca2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:AUUGGRYYUL3DI2A65J5R2Y22TL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Internal Language Model Estimation Through Explicit Context Vector Learning for Attention-based Encoder-decoder ASR","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.SD"],"primary_cat":"eess.AS","authors_text":"Haihua Xu, Rao Ma, Weibin Zhang, Yi He, Yufei Liu, Zejun Ma","submitted_at":"2022-01-26T07:47:27Z","abstract_excerpt":"An end-to-end (E2E) ASR model implicitly learns a prior Internal Language Model (ILM) from the training transcripts. To fuse an external LM using Bayes posterior theory, the log likelihood produced by the ILM has to be accurately estimated and subtracted. In this paper we propose two novel approaches to estimate the ILM based on Listen-Attend-Spell (LAS) framework. The first method is to replace the context vector of the LAS decoder at every time step with a vector that is learned with training transcripts. Furthermore, we propose another method that uses a lightweight feed-forward network to "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.11627","kind":"arxiv","version":3},"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/2201.11627/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":"2026-07-05T05:12:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kgNp/aMTRRAn6r05n9ldFD0O1aCnB+KGqqbqRYn7cSSHcc4bI7eIhMt47mDRwx+RY1fiTrTSvlaTcgyA14tnBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-15T23:48:27.444987Z"},"content_sha256":"61e76c9393e6b1b208614c48256f69746c8750ae51858479b2636e189ec69070","schema_version":"1.0","event_id":"sha256:61e76c9393e6b1b208614c48256f69746c8750ae51858479b2636e189ec69070"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AUUGGRYYUL3DI2A65J5R2Y22TL/bundle.json","state_url":"https://pith.science/pith/AUUGGRYYUL3DI2A65J5R2Y22TL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AUUGGRYYUL3DI2A65J5R2Y22TL/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":"2026-07-15T23:48:27Z","links":{"resolver":"https://pith.science/pith/AUUGGRYYUL3DI2A65J5R2Y22TL","bundle":"https://pith.science/pith/AUUGGRYYUL3DI2A65J5R2Y22TL/bundle.json","state":"https://pith.science/pith/AUUGGRYYUL3DI2A65J5R2Y22TL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AUUGGRYYUL3DI2A65J5R2Y22TL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:AUUGGRYYUL3DI2A65J5R2Y22TL","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":"fa8f0783d5c005a47fc4b3a275596c0f244984e09b5679f0ba4c666beb3c8527","cross_cats_sorted":["cs.AI","cs.SD"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-01-26T07:47:27Z","title_canon_sha256":"82e15d21826ebc9143f3b60b307fb6eaffa02554f72ffa36976a65112668d032"},"schema_version":"1.0","source":{"id":"2201.11627","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2201.11627","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"arxiv_version","alias_value":"2201.11627v3","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2201.11627","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"pith_short_12","alias_value":"AUUGGRYYUL3D","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"pith_short_16","alias_value":"AUUGGRYYUL3DI2A6","created_at":"2026-07-05T05:12:27Z"},{"alias_kind":"pith_short_8","alias_value":"AUUGGRYY","created_at":"2026-07-05T05:12:27Z"}],"graph_snapshots":[{"event_id":"sha256:61e76c9393e6b1b208614c48256f69746c8750ae51858479b2636e189ec69070","target":"graph","created_at":"2026-07-05T05:12:27Z","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/2201.11627/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"An end-to-end (E2E) ASR model implicitly learns a prior Internal Language Model (ILM) from the training transcripts. To fuse an external LM using Bayes posterior theory, the log likelihood produced by the ILM has to be accurately estimated and subtracted. In this paper we propose two novel approaches to estimate the ILM based on Listen-Attend-Spell (LAS) framework. The first method is to replace the context vector of the LAS decoder at every time step with a vector that is learned with training transcripts. Furthermore, we propose another method that uses a lightweight feed-forward network to ","authors_text":"Haihua Xu, Rao Ma, Weibin Zhang, Yi He, Yufei Liu, Zejun Ma","cross_cats":["cs.AI","cs.SD"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-01-26T07:47:27Z","title":"Internal Language Model Estimation Through Explicit Context Vector Learning for Attention-based Encoder-decoder ASR"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2201.11627","kind":"arxiv","version":3},"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":"sha256:6b5805a5e35b96dad0b0be757102372e7c7efa924261b96ed4c168d462b87ca2","target":"record","created_at":"2026-07-05T05:12:27Z","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":"fa8f0783d5c005a47fc4b3a275596c0f244984e09b5679f0ba4c666beb3c8527","cross_cats_sorted":["cs.AI","cs.SD"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-01-26T07:47:27Z","title_canon_sha256":"82e15d21826ebc9143f3b60b307fb6eaffa02554f72ffa36976a65112668d032"},"schema_version":"1.0","source":{"id":"2201.11627","kind":"arxiv","version":3}},"canonical_sha256":"0528634718a2f634681eea7b1d635a9ad8ee400c152ed5face1d3807ddee763e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0528634718a2f634681eea7b1d635a9ad8ee400c152ed5face1d3807ddee763e","first_computed_at":"2026-07-05T05:12:27.198099Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:12:27.198099Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/YA4pyb3MdUwmiDCNfU+Ek8xPiYHr0PM6PV49WbcwBCZuK0rvoaRsS4qFJ8Ou4uqyPJT2MjoOTcr3TfJL+omAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T05:12:27.198609Z","signed_message":"canonical_sha256_bytes"},"source_id":"2201.11627","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6b5805a5e35b96dad0b0be757102372e7c7efa924261b96ed4c168d462b87ca2","sha256:61e76c9393e6b1b208614c48256f69746c8750ae51858479b2636e189ec69070"],"state_sha256":"0a77977265749cff54edcb0ea83df414eb2da61d195b9274ce448ea0f8bd7359"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nqHHpkss0sYff3hHuQSor7U/PV4SQJj4rw7fOUjwS1u7ST+oDwCPB4zgKwSu3vXD7mdlrWvg7ihav19MSvEaBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-15T23:48:27.447522Z","bundle_sha256":"dabe8fb6706535a2e15d61102237d7b09ab4bf02a5a48a9e4e8666d8f8a33548"}}