{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:NSSGHJLU4TUSRWTEBQCTX5X4BD","short_pith_number":"pith:NSSGHJLU","schema_version":"1.0","canonical_sha256":"6ca463a574e4e928da640c053bf6fc08c19b792fb40ceb1eeebfc33aacaab551","source":{"kind":"arxiv","id":"2204.07837","version":2},"attestation_state":"computed","paper":{"title":"BLISS: Robust Sequence-to-Sequence Learning via Self-Supervised Input Representation","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dacheng Tao, Dazhao Cheng, Liang Ding, Min Zhang, Xuebo Liu, Zheng Zhang","submitted_at":"2022-04-16T16:19:47Z","abstract_excerpt":"Data augmentations (DA) are the cores to achieving robust sequence-to-sequence learning on various natural language processing (NLP) tasks. However, most of the DA approaches force the decoder to make predictions conditioned on the perturbed input representation, underutilizing supervised information provided by perturbed input. In this work, we propose a framework-level robust sequence-to-sequence learning approach, named BLISS, via self-supervised input representation, which has the great potential to complement the data-level augmentation approaches. The key idea is to supervise the sequenc"},"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":"2204.07837","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2022-04-16T16:19:47Z","cross_cats_sorted":[],"title_canon_sha256":"1272cc3063d3175ae6060ce64e48397a645b3e5823e7b16bc17b1be6e643f64c","abstract_canon_sha256":"70929624b73356caf7f456753ae6e15ef18fdca3f59a6d89802d57533ec7cc8c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:17:22.417374Z","signature_b64":"0UZOWlAZlvfHYCHItm38tPxQ1wWJsAnbdjzZVcJ62kSUXFZV3QJm6gqqG4wTsMmlncGbrySx6yzBUAtIskrUCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6ca463a574e4e928da640c053bf6fc08c19b792fb40ceb1eeebfc33aacaab551","last_reissued_at":"2026-07-05T04:17:22.416857Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:17:22.416857Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BLISS: Robust Sequence-to-Sequence Learning via Self-Supervised Input Representation","license":"http://creativecommons.org/publicdomain/zero/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dacheng Tao, Dazhao Cheng, Liang Ding, Min Zhang, Xuebo Liu, Zheng Zhang","submitted_at":"2022-04-16T16:19:47Z","abstract_excerpt":"Data augmentations (DA) are the cores to achieving robust sequence-to-sequence learning on various natural language processing (NLP) tasks. However, most of the DA approaches force the decoder to make predictions conditioned on the perturbed input representation, underutilizing supervised information provided by perturbed input. In this work, we propose a framework-level robust sequence-to-sequence learning approach, named BLISS, via self-supervised input representation, which has the great potential to complement the data-level augmentation approaches. The key idea is to supervise the sequenc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2204.07837","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/2204.07837/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":"2204.07837","created_at":"2026-07-05T04:17:22.416944+00:00"},{"alias_kind":"arxiv_version","alias_value":"2204.07837v2","created_at":"2026-07-05T04:17:22.416944+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2204.07837","created_at":"2026-07-05T04:17:22.416944+00:00"},{"alias_kind":"pith_short_12","alias_value":"NSSGHJLU4TUS","created_at":"2026-07-05T04:17:22.416944+00:00"},{"alias_kind":"pith_short_16","alias_value":"NSSGHJLU4TUSRWTE","created_at":"2026-07-05T04:17:22.416944+00:00"},{"alias_kind":"pith_short_8","alias_value":"NSSGHJLU","created_at":"2026-07-05T04:17:22.416944+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.22175","citing_title":"MPipeMoE: Memory Efficient MoE for Pre-trained Models with Adaptive Pipeline Parallelism","ref_index":11,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/NSSGHJLU4TUSRWTEBQCTX5X4BD","json":"https://pith.science/pith/NSSGHJLU4TUSRWTEBQCTX5X4BD.json","graph_json":"https://pith.science/api/pith-number/NSSGHJLU4TUSRWTEBQCTX5X4BD/graph.json","events_json":"https://pith.science/api/pith-number/NSSGHJLU4TUSRWTEBQCTX5X4BD/events.json","paper":"https://pith.science/paper/NSSGHJLU"},"agent_actions":{"view_html":"https://pith.science/pith/NSSGHJLU4TUSRWTEBQCTX5X4BD","download_json":"https://pith.science/pith/NSSGHJLU4TUSRWTEBQCTX5X4BD.json","view_paper":"https://pith.science/paper/NSSGHJLU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2204.07837&json=true","fetch_graph":"https://pith.science/api/pith-number/NSSGHJLU4TUSRWTEBQCTX5X4BD/graph.json","fetch_events":"https://pith.science/api/pith-number/NSSGHJLU4TUSRWTEBQCTX5X4BD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NSSGHJLU4TUSRWTEBQCTX5X4BD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NSSGHJLU4TUSRWTEBQCTX5X4BD/action/storage_attestation","attest_author":"https://pith.science/pith/NSSGHJLU4TUSRWTEBQCTX5X4BD/action/author_attestation","sign_citation":"https://pith.science/pith/NSSGHJLU4TUSRWTEBQCTX5X4BD/action/citation_signature","submit_replication":"https://pith.science/pith/NSSGHJLU4TUSRWTEBQCTX5X4BD/action/replication_record"}},"created_at":"2026-07-05T04:17:22.416944+00:00","updated_at":"2026-07-05T04:17:22.416944+00:00"}