{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:4VTA7EU6B76SNIYCAPDRJDNQO6","short_pith_number":"pith:4VTA7EU6","schema_version":"1.0","canonical_sha256":"e5660f929e0ffd26a30203c7148db0779401c75a319f3680ec69e43e235396f5","source":{"kind":"arxiv","id":"2108.11846","version":1},"attestation_state":"computed","paper":{"title":"Alleviating Exposure Bias via Contrastive Learning for Abstractive Text Summarization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Shichao Sun, Wenjie Li","submitted_at":"2021-08-26T15:14:44Z","abstract_excerpt":"Encoder-decoder models have achieved remarkable success in abstractive text summarization, which aims to compress one or more documents into a shorter version without the loss of the essential content. Unfortunately, these models mostly suffer a discrepancy between training and inference, i.e., the exposure bias problem. During the training stage, with teacher forcing these models are optimized to maximize the likelihood of the gold summary given the gold summary tokens as input to the decoder, while at inference the given tokens are replaced by the generated tokens. Consequently, low-quality "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2108.11846","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2021-08-26T15:14:44Z","cross_cats_sorted":[],"title_canon_sha256":"2a09fd6ae8dc4224d57c8d21152c80366825bc1155b3bb5ea3142509f7dc7acf","abstract_canon_sha256":"f532f0864bbda53ce52e78da4a838dae049c38743b1d76a5e841b30c053f430b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:09:10.361746Z","signature_b64":"06h9wFnzS3nfg8s5pLar44U+mjJ1nGqFhwjgmvhMaGwFAVFN63PvBRqZ63htsZP39Sw2tg1gqVWGBLzbC04UDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e5660f929e0ffd26a30203c7148db0779401c75a319f3680ec69e43e235396f5","last_reissued_at":"2026-07-05T03:09:10.361345Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:09:10.361345Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Alleviating Exposure Bias via Contrastive Learning for Abstractive Text Summarization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Shichao Sun, Wenjie Li","submitted_at":"2021-08-26T15:14:44Z","abstract_excerpt":"Encoder-decoder models have achieved remarkable success in abstractive text summarization, which aims to compress one or more documents into a shorter version without the loss of the essential content. Unfortunately, these models mostly suffer a discrepancy between training and inference, i.e., the exposure bias problem. During the training stage, with teacher forcing these models are optimized to maximize the likelihood of the gold summary given the gold summary tokens as input to the decoder, while at inference the given tokens are replaced by the generated tokens. Consequently, low-quality "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.11846","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/2108.11846/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2108.11846","created_at":"2026-07-05T03:09:10.361401+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.11846v1","created_at":"2026-07-05T03:09:10.361401+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.11846","created_at":"2026-07-05T03:09:10.361401+00:00"},{"alias_kind":"pith_short_12","alias_value":"4VTA7EU6B76S","created_at":"2026-07-05T03:09:10.361401+00:00"},{"alias_kind":"pith_short_16","alias_value":"4VTA7EU6B76SNIYC","created_at":"2026-07-05T03:09:10.361401+00:00"},{"alias_kind":"pith_short_8","alias_value":"4VTA7EU6","created_at":"2026-07-05T03:09:10.361401+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4VTA7EU6B76SNIYCAPDRJDNQO6","json":"https://pith.science/pith/4VTA7EU6B76SNIYCAPDRJDNQO6.json","graph_json":"https://pith.science/api/pith-number/4VTA7EU6B76SNIYCAPDRJDNQO6/graph.json","events_json":"https://pith.science/api/pith-number/4VTA7EU6B76SNIYCAPDRJDNQO6/events.json","paper":"https://pith.science/paper/4VTA7EU6"},"agent_actions":{"view_html":"https://pith.science/pith/4VTA7EU6B76SNIYCAPDRJDNQO6","download_json":"https://pith.science/pith/4VTA7EU6B76SNIYCAPDRJDNQO6.json","view_paper":"https://pith.science/paper/4VTA7EU6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.11846&json=true","fetch_graph":"https://pith.science/api/pith-number/4VTA7EU6B76SNIYCAPDRJDNQO6/graph.json","fetch_events":"https://pith.science/api/pith-number/4VTA7EU6B76SNIYCAPDRJDNQO6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4VTA7EU6B76SNIYCAPDRJDNQO6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4VTA7EU6B76SNIYCAPDRJDNQO6/action/storage_attestation","attest_author":"https://pith.science/pith/4VTA7EU6B76SNIYCAPDRJDNQO6/action/author_attestation","sign_citation":"https://pith.science/pith/4VTA7EU6B76SNIYCAPDRJDNQO6/action/citation_signature","submit_replication":"https://pith.science/pith/4VTA7EU6B76SNIYCAPDRJDNQO6/action/replication_record"}},"created_at":"2026-07-05T03:09:10.361401+00:00","updated_at":"2026-07-05T03:09:10.361401+00:00"}