{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:6E6CMEMIL2IOEDBR4DNNWUL535","short_pith_number":"pith:6E6CMEMI","canonical_record":{"source":{"id":"2106.03441","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-07T09:18:21Z","cross_cats_sorted":[],"title_canon_sha256":"3f4845d1e8fc075e243f442962b2f435517a19801961bdc283ee3bd85b481f2c","abstract_canon_sha256":"d4681e3230a2c69a30a081a629ace9bdf085fb0aee831c018002999bceabc661"},"schema_version":"1.0"},"canonical_sha256":"f13c2611885e90e20c31e0dadb517ddf7e7c062beae4e2cca094fe88c1179ec3","source":{"kind":"arxiv","id":"2106.03441","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.03441","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"arxiv_version","alias_value":"2106.03441v3","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.03441","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"pith_short_12","alias_value":"6E6CMEMIL2IO","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"pith_short_16","alias_value":"6E6CMEMIL2IOEDBR","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"pith_short_8","alias_value":"6E6CMEMI","created_at":"2026-07-05T04:00:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:6E6CMEMIL2IOEDBR4DNNWUL535","target":"record","payload":{"canonical_record":{"source":{"id":"2106.03441","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-07T09:18:21Z","cross_cats_sorted":[],"title_canon_sha256":"3f4845d1e8fc075e243f442962b2f435517a19801961bdc283ee3bd85b481f2c","abstract_canon_sha256":"d4681e3230a2c69a30a081a629ace9bdf085fb0aee831c018002999bceabc661"},"schema_version":"1.0"},"canonical_sha256":"f13c2611885e90e20c31e0dadb517ddf7e7c062beae4e2cca094fe88c1179ec3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:00:53.610975Z","signature_b64":"/3Gy3o7xM5uiP2ub9yagAtZtT+87b8e/OjvmVTpWQJXbzmFVXHSK66jTdmRqmyKHNCjDYhD8Tue/CWAZo7ZzDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f13c2611885e90e20c31e0dadb517ddf7e7c062beae4e2cca094fe88c1179ec3","last_reissued_at":"2026-07-05T04:00:53.610518Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:00:53.610518Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2106.03441","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":"2026-07-05T04:00:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4BjSQrkYANCaGHBqI1JPb1YOa7Qs61OcvM9Dj8Mhcg8UnqesBkCOK39VK/+uxTkNVUeeZT2DtMzL77TPlGBcCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T18:39:45.547446Z"},"content_sha256":"dafc39504e657fc4fd9562970f0a620e2719a1d7ae932add295dd9218ac2084b","schema_version":"1.0","event_id":"sha256:dafc39504e657fc4fd9562970f0a620e2719a1d7ae932add295dd9218ac2084b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:6E6CMEMIL2IOEDBR4DNNWUL535","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Attention Temperature Matters in Abstractive Summarization Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Furu Wei, Hangbo Bao, Shengqiang Zhang, Xingxing Zhang","submitted_at":"2021-06-07T09:18:21Z","abstract_excerpt":"Recent progress of abstractive text summarization largely relies on large pre-trained sequence-to-sequence Transformer models, which are computationally expensive. This paper aims to distill these large models into smaller ones for faster inference and minimal performance loss. Pseudo-labeling based methods are popular in sequence-to-sequence model distillation. In this paper, we find simply manipulating attention temperatures in Transformers can make pseudo labels easier to learn for student models. Our experiments on three summarization datasets show our proposed method consistently improves"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.03441","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/2106.03441/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":"2026-07-05T04:00:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vLFbO+q658DzMm8mUr7fGZpWM2uiTtk2CB/6UiOD0zZns0cKLFldhiAb7U1TH95cmut9Sb8EHVpXHEXkQlexCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T18:39:45.547834Z"},"content_sha256":"407acaba784eb7fdead9f24def99067569335f338ca46532c7b9b5dc56267de8","schema_version":"1.0","event_id":"sha256:407acaba784eb7fdead9f24def99067569335f338ca46532c7b9b5dc56267de8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6E6CMEMIL2IOEDBR4DNNWUL535/bundle.json","state_url":"https://pith.science/pith/6E6CMEMIL2IOEDBR4DNNWUL535/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6E6CMEMIL2IOEDBR4DNNWUL535/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":"2026-07-06T18:39:45Z","links":{"resolver":"https://pith.science/pith/6E6CMEMIL2IOEDBR4DNNWUL535","bundle":"https://pith.science/pith/6E6CMEMIL2IOEDBR4DNNWUL535/bundle.json","state":"https://pith.science/pith/6E6CMEMIL2IOEDBR4DNNWUL535/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6E6CMEMIL2IOEDBR4DNNWUL535/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:6E6CMEMIL2IOEDBR4DNNWUL535","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":"d4681e3230a2c69a30a081a629ace9bdf085fb0aee831c018002999bceabc661","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-07T09:18:21Z","title_canon_sha256":"3f4845d1e8fc075e243f442962b2f435517a19801961bdc283ee3bd85b481f2c"},"schema_version":"1.0","source":{"id":"2106.03441","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2106.03441","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"arxiv_version","alias_value":"2106.03441v3","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2106.03441","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"pith_short_12","alias_value":"6E6CMEMIL2IO","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"pith_short_16","alias_value":"6E6CMEMIL2IOEDBR","created_at":"2026-07-05T04:00:53Z"},{"alias_kind":"pith_short_8","alias_value":"6E6CMEMI","created_at":"2026-07-05T04:00:53Z"}],"graph_snapshots":[{"event_id":"sha256:407acaba784eb7fdead9f24def99067569335f338ca46532c7b9b5dc56267de8","target":"graph","created_at":"2026-07-05T04:00:53Z","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/2106.03441/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent progress of abstractive text summarization largely relies on large pre-trained sequence-to-sequence Transformer models, which are computationally expensive. This paper aims to distill these large models into smaller ones for faster inference and minimal performance loss. Pseudo-labeling based methods are popular in sequence-to-sequence model distillation. In this paper, we find simply manipulating attention temperatures in Transformers can make pseudo labels easier to learn for student models. Our experiments on three summarization datasets show our proposed method consistently improves","authors_text":"Furu Wei, Hangbo Bao, Shengqiang Zhang, Xingxing Zhang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-07T09:18:21Z","title":"Attention Temperature Matters in Abstractive Summarization Distillation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2106.03441","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:dafc39504e657fc4fd9562970f0a620e2719a1d7ae932add295dd9218ac2084b","target":"record","created_at":"2026-07-05T04:00:53Z","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":"d4681e3230a2c69a30a081a629ace9bdf085fb0aee831c018002999bceabc661","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-06-07T09:18:21Z","title_canon_sha256":"3f4845d1e8fc075e243f442962b2f435517a19801961bdc283ee3bd85b481f2c"},"schema_version":"1.0","source":{"id":"2106.03441","kind":"arxiv","version":3}},"canonical_sha256":"f13c2611885e90e20c31e0dadb517ddf7e7c062beae4e2cca094fe88c1179ec3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f13c2611885e90e20c31e0dadb517ddf7e7c062beae4e2cca094fe88c1179ec3","first_computed_at":"2026-07-05T04:00:53.610518Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:00:53.610518Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/3Gy3o7xM5uiP2ub9yagAtZtT+87b8e/OjvmVTpWQJXbzmFVXHSK66jTdmRqmyKHNCjDYhD8Tue/CWAZo7ZzDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T04:00:53.610975Z","signed_message":"canonical_sha256_bytes"},"source_id":"2106.03441","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dafc39504e657fc4fd9562970f0a620e2719a1d7ae932add295dd9218ac2084b","sha256:407acaba784eb7fdead9f24def99067569335f338ca46532c7b9b5dc56267de8"],"state_sha256":"ac462e670b544dc00b171fc835f515afc0f0b95dec69b973c85e2a435bcc73da"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"7cHr2vcfZPNfGDyNdLR3VcYzt0LA1C4rPvOXalNFBuNIu68pXNOQ9jIEIlTHYZmKzOxwysdvnwjAGaHVgSkzBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T18:39:45.550031Z","bundle_sha256":"9e7bb32c3b30c546cdef94de9a388ab952d5c0086c34de6f94c05a7379522a37"}}