{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:NO2EN4U2UQS3B7BDB3UM4NAHKB","short_pith_number":"pith:NO2EN4U2","schema_version":"1.0","canonical_sha256":"6bb446f29aa425b0fc230ee8ce3407504b1d5cc0185b493b0136f07e220e93e2","source":{"kind":"arxiv","id":"2208.14641","version":1},"attestation_state":"computed","paper":{"title":"Generating Intermediate Steps for NLI with Next-Step Supervision","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Deepanway Ghosal, Monojit Choudhury, Somak Aditya","submitted_at":"2022-08-31T05:25:33Z","abstract_excerpt":"The Natural Language Inference (NLI) task often requires reasoning over multiple steps to reach the conclusion. While the necessity of generating such intermediate steps (instead of a summary explanation) has gained popular support, it is unclear how to generate such steps without complete end-to-end supervision and how such generated steps can be further utilized. In this work, we train a sequence-to-sequence model to generate only the next step given an NLI premise and hypothesis pair (and previous steps); then enhance it with external knowledge and symbolic search to generate intermediate s"},"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":"2208.14641","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2022-08-31T05:25:33Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3afbdc5ab375e63c911e28a4bbd225dcee63542bdc80f62a73be3c021670be52","abstract_canon_sha256":"4244f476fdc9c1ecc8b077dc5a25faf120183d1bb63e33b30799744d09ea6182"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:53:32.553588Z","signature_b64":"P52YSGcyzHrvUUJgZ4Qg2A0BG5fMavXKuIC99ohkSRrDtWeU03gdMeEihbinKoNnSlRbWpVFCr12x/Nhu0oGDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6bb446f29aa425b0fc230ee8ce3407504b1d5cc0185b493b0136f07e220e93e2","last_reissued_at":"2026-07-05T04:53:32.553226Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:53:32.553226Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generating Intermediate Steps for NLI with Next-Step Supervision","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Deepanway Ghosal, Monojit Choudhury, Somak Aditya","submitted_at":"2022-08-31T05:25:33Z","abstract_excerpt":"The Natural Language Inference (NLI) task often requires reasoning over multiple steps to reach the conclusion. While the necessity of generating such intermediate steps (instead of a summary explanation) has gained popular support, it is unclear how to generate such steps without complete end-to-end supervision and how such generated steps can be further utilized. In this work, we train a sequence-to-sequence model to generate only the next step given an NLI premise and hypothesis pair (and previous steps); then enhance it with external knowledge and symbolic search to generate intermediate s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2208.14641","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/2208.14641/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":"2208.14641","created_at":"2026-07-05T04:53:32.553279+00:00"},{"alias_kind":"arxiv_version","alias_value":"2208.14641v1","created_at":"2026-07-05T04:53:32.553279+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2208.14641","created_at":"2026-07-05T04:53:32.553279+00:00"},{"alias_kind":"pith_short_12","alias_value":"NO2EN4U2UQS3","created_at":"2026-07-05T04:53:32.553279+00:00"},{"alias_kind":"pith_short_16","alias_value":"NO2EN4U2UQS3B7BD","created_at":"2026-07-05T04:53:32.553279+00:00"},{"alias_kind":"pith_short_8","alias_value":"NO2EN4U2","created_at":"2026-07-05T04:53:32.553279+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/NO2EN4U2UQS3B7BDB3UM4NAHKB","json":"https://pith.science/pith/NO2EN4U2UQS3B7BDB3UM4NAHKB.json","graph_json":"https://pith.science/api/pith-number/NO2EN4U2UQS3B7BDB3UM4NAHKB/graph.json","events_json":"https://pith.science/api/pith-number/NO2EN4U2UQS3B7BDB3UM4NAHKB/events.json","paper":"https://pith.science/paper/NO2EN4U2"},"agent_actions":{"view_html":"https://pith.science/pith/NO2EN4U2UQS3B7BDB3UM4NAHKB","download_json":"https://pith.science/pith/NO2EN4U2UQS3B7BDB3UM4NAHKB.json","view_paper":"https://pith.science/paper/NO2EN4U2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2208.14641&json=true","fetch_graph":"https://pith.science/api/pith-number/NO2EN4U2UQS3B7BDB3UM4NAHKB/graph.json","fetch_events":"https://pith.science/api/pith-number/NO2EN4U2UQS3B7BDB3UM4NAHKB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NO2EN4U2UQS3B7BDB3UM4NAHKB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NO2EN4U2UQS3B7BDB3UM4NAHKB/action/storage_attestation","attest_author":"https://pith.science/pith/NO2EN4U2UQS3B7BDB3UM4NAHKB/action/author_attestation","sign_citation":"https://pith.science/pith/NO2EN4U2UQS3B7BDB3UM4NAHKB/action/citation_signature","submit_replication":"https://pith.science/pith/NO2EN4U2UQS3B7BDB3UM4NAHKB/action/replication_record"}},"created_at":"2026-07-05T04:53:32.553279+00:00","updated_at":"2026-07-05T04:53:32.553279+00:00"}