{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:NYHTCIAQSDUVVQI35GX5XU6S2A","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":"d6a37fcd91cabeeef641ef852007f123c7acb1d210b2ed7666af2752503a04c3","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-06T00:20:11Z","title_canon_sha256":"b18ba93c3c120e15394466e60d2b07c5bdb83f589ccee13ab376428262a6aeb9"},"schema_version":"1.0","source":{"id":"2311.02805","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.02805","created_at":"2026-07-05T08:21:43Z"},{"alias_kind":"arxiv_version","alias_value":"2311.02805v2","created_at":"2026-07-05T08:21:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.02805","created_at":"2026-07-05T08:21:43Z"},{"alias_kind":"pith_short_12","alias_value":"NYHTCIAQSDUV","created_at":"2026-07-05T08:21:43Z"},{"alias_kind":"pith_short_16","alias_value":"NYHTCIAQSDUVVQI3","created_at":"2026-07-05T08:21:43Z"},{"alias_kind":"pith_short_8","alias_value":"NYHTCIAQ","created_at":"2026-07-05T08:21:43Z"}],"graph_snapshots":[{"event_id":"sha256:11229a41457b58985a81c68ea1be367f503259f7f48ee4c502b0354229f51674","target":"graph","created_at":"2026-07-05T08:21:43Z","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/2311.02805/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LMs) are capable of generating free-text rationales to aid question answering. However, prior work 1) suggests that useful self-rationalization is emergent only at significant scales (e.g., 175B parameter GPT-3); and 2) focuses largely on downstream performance, ignoring the semantics of the rationales themselves, e.g., are they faithful, true, and helpful for humans? In this work, we enable small-scale LMs (approx. 200x smaller than GPT-3) to generate rationales that not only improve downstream task performance, but are also more plausible, consistent, and diverse, asse","authors_text":"Aaron Chan, Brihi Joshi, Jack Hessel, Liunian Harold Li, Sahana Ramnath, Skyler Hallinan, Xiang Ren, Ximing Lu, Yejin Choi","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-06T00:20:11Z","title":"Tailoring Self-Rationalizers with Multi-Reward Distillation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.02805","kind":"arxiv","version":2},"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":"sha256:79ef4febc102ecdc8da7c67d95eef9fb1ed76a47431595b8954a5d97df808020","target":"record","created_at":"2026-07-05T08:21:43Z","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":"d6a37fcd91cabeeef641ef852007f123c7acb1d210b2ed7666af2752503a04c3","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-06T00:20:11Z","title_canon_sha256":"b18ba93c3c120e15394466e60d2b07c5bdb83f589ccee13ab376428262a6aeb9"},"schema_version":"1.0","source":{"id":"2311.02805","kind":"arxiv","version":2}},"canonical_sha256":"6e0f31201090e95ac11be9afdbd3d2d03e07a094aae1efef9e31dceb9d070939","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6e0f31201090e95ac11be9afdbd3d2d03e07a094aae1efef9e31dceb9d070939","first_computed_at":"2026-07-05T08:21:43.347149Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:21:43.347149Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1PLIy6Dycanomg/1o1OiChlKljQXtNLwJOmQ5Vbr8sdSY89CdhaGTVVn2tBC0RY3qLYa6z0dn6UphGBMn3dpAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:21:43.347641Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.02805","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:79ef4febc102ecdc8da7c67d95eef9fb1ed76a47431595b8954a5d97df808020","sha256:11229a41457b58985a81c68ea1be367f503259f7f48ee4c502b0354229f51674"],"state_sha256":"24135a9f6d2d028ba497237cefbb8e6d238401b2b4605d2d88d075a65b86c0d0"}