{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:GE74KNB3N3XDDIQWGISXXQNYYV","short_pith_number":"pith:GE74KNB3","canonical_record":{"source":{"id":"2205.03401","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-05-06T17:57:58Z","cross_cats_sorted":[],"title_canon_sha256":"04d7f84b83b48b4a48b7b16282d8fc1ecb6432aec1a1bd0474dd6d6eea9e5cca","abstract_canon_sha256":"c9c2b61d13eb298f47400fae6e68b86e1ca13dcb153361e7cb475f42820f59be"},"schema_version":"1.0"},"canonical_sha256":"313fc5343b6eee31a21632257bc1b8c55aa0e5153224e76067e2a319f6e0ae7c","source":{"kind":"arxiv","id":"2205.03401","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2205.03401","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"arxiv_version","alias_value":"2205.03401v2","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.03401","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"pith_short_12","alias_value":"GE74KNB3N3XD","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"pith_short_16","alias_value":"GE74KNB3N3XDDIQW","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"pith_short_8","alias_value":"GE74KNB3","created_at":"2026-07-05T05:06:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:GE74KNB3N3XDDIQWGISXXQNYYV","target":"record","payload":{"canonical_record":{"source":{"id":"2205.03401","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-05-06T17:57:58Z","cross_cats_sorted":[],"title_canon_sha256":"04d7f84b83b48b4a48b7b16282d8fc1ecb6432aec1a1bd0474dd6d6eea9e5cca","abstract_canon_sha256":"c9c2b61d13eb298f47400fae6e68b86e1ca13dcb153361e7cb475f42820f59be"},"schema_version":"1.0"},"canonical_sha256":"313fc5343b6eee31a21632257bc1b8c55aa0e5153224e76067e2a319f6e0ae7c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:06:11.423750Z","signature_b64":"Q11MKIgq4rxaFwgszdwZP9ZKFGy1jtdfLZGYOxHNeLAfiLkWUfzJK8PuBEz4GLyi+3IGyWNcemAJtf9AQnGOBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"313fc5343b6eee31a21632257bc1b8c55aa0e5153224e76067e2a319f6e0ae7c","last_reissued_at":"2026-07-05T05:06:11.423345Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:06:11.423345Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2205.03401","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:06:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zUunVSR67tmi3mskW8tkoIkeLdtl0ZscMECYXU7Gj/hNn4/WqzWnHdiB6Q9qKmoVfhWKezn/Bz5eyX+locRuBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T13:38:11.450429Z"},"content_sha256":"10f3c1636afc1ecb848c05d2fcba618bf23e56cd447df076f069ab54131b7b79","schema_version":"1.0","event_id":"sha256:10f3c1636afc1ecb848c05d2fcba618bf23e56cd447df076f069ab54131b7b79"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:GE74KNB3N3XDDIQWGISXXQNYYV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Unreliability of Explanations in Few-shot Prompting for Textual Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Greg Durrett, Xi Ye","submitted_at":"2022-05-06T17:57:58Z","abstract_excerpt":"Does prompting a large language model (LLM) like GPT-3 with explanations improve in-context learning? We study this question on two NLP tasks that involve reasoning over text, namely question answering and natural language inference. We test the performance of four LLMs on three textual reasoning datasets using prompts that include explanations in multiple different styles. For these tasks, we find that including explanations in the prompts for OPT, GPT-3 (davinci), and InstructGPT (text-davinci-001) only yields small to moderate accuracy improvements over standard few-show learning. However, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.03401","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/2205.03401/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":"2026-07-05T05:06:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nFVk1AgiWZi5qxpgWqcjDhu7uVl3V0dhCwqIi3b9lRTcz1qxORB5HbT/jeF2ZkJl6B6QE1jVU3yqH8t8dSwQDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T13:38:11.451058Z"},"content_sha256":"2b983cda62eabd90ffe674f296926cff202d00f78f3da1b882f09c909f4a6ef0","schema_version":"1.0","event_id":"sha256:2b983cda62eabd90ffe674f296926cff202d00f78f3da1b882f09c909f4a6ef0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GE74KNB3N3XDDIQWGISXXQNYYV/bundle.json","state_url":"https://pith.science/pith/GE74KNB3N3XDDIQWGISXXQNYYV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GE74KNB3N3XDDIQWGISXXQNYYV/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":"2026-08-14T13:38:11Z","links":{"resolver":"https://pith.science/pith/GE74KNB3N3XDDIQWGISXXQNYYV","bundle":"https://pith.science/pith/GE74KNB3N3XDDIQWGISXXQNYYV/bundle.json","state":"https://pith.science/pith/GE74KNB3N3XDDIQWGISXXQNYYV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GE74KNB3N3XDDIQWGISXXQNYYV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:GE74KNB3N3XDDIQWGISXXQNYYV","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":"c9c2b61d13eb298f47400fae6e68b86e1ca13dcb153361e7cb475f42820f59be","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-05-06T17:57:58Z","title_canon_sha256":"04d7f84b83b48b4a48b7b16282d8fc1ecb6432aec1a1bd0474dd6d6eea9e5cca"},"schema_version":"1.0","source":{"id":"2205.03401","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2205.03401","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"arxiv_version","alias_value":"2205.03401v2","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.03401","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"pith_short_12","alias_value":"GE74KNB3N3XD","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"pith_short_16","alias_value":"GE74KNB3N3XDDIQW","created_at":"2026-07-05T05:06:11Z"},{"alias_kind":"pith_short_8","alias_value":"GE74KNB3","created_at":"2026-07-05T05:06:11Z"}],"graph_snapshots":[{"event_id":"sha256:2b983cda62eabd90ffe674f296926cff202d00f78f3da1b882f09c909f4a6ef0","target":"graph","created_at":"2026-07-05T05:06:11Z","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/2205.03401/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Does prompting a large language model (LLM) like GPT-3 with explanations improve in-context learning? We study this question on two NLP tasks that involve reasoning over text, namely question answering and natural language inference. We test the performance of four LLMs on three textual reasoning datasets using prompts that include explanations in multiple different styles. For these tasks, we find that including explanations in the prompts for OPT, GPT-3 (davinci), and InstructGPT (text-davinci-001) only yields small to moderate accuracy improvements over standard few-show learning. However, ","authors_text":"Greg Durrett, Xi Ye","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-05-06T17:57:58Z","title":"The Unreliability of Explanations in Few-shot Prompting for Textual Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.03401","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":"sha256:10f3c1636afc1ecb848c05d2fcba618bf23e56cd447df076f069ab54131b7b79","target":"record","created_at":"2026-07-05T05:06:11Z","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":"c9c2b61d13eb298f47400fae6e68b86e1ca13dcb153361e7cb475f42820f59be","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2022-05-06T17:57:58Z","title_canon_sha256":"04d7f84b83b48b4a48b7b16282d8fc1ecb6432aec1a1bd0474dd6d6eea9e5cca"},"schema_version":"1.0","source":{"id":"2205.03401","kind":"arxiv","version":2}},"canonical_sha256":"313fc5343b6eee31a21632257bc1b8c55aa0e5153224e76067e2a319f6e0ae7c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"313fc5343b6eee31a21632257bc1b8c55aa0e5153224e76067e2a319f6e0ae7c","first_computed_at":"2026-07-05T05:06:11.423345Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:06:11.423345Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Q11MKIgq4rxaFwgszdwZP9ZKFGy1jtdfLZGYOxHNeLAfiLkWUfzJK8PuBEz4GLyi+3IGyWNcemAJtf9AQnGOBg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:06:11.423750Z","signed_message":"canonical_sha256_bytes"},"source_id":"2205.03401","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:10f3c1636afc1ecb848c05d2fcba618bf23e56cd447df076f069ab54131b7b79","sha256:2b983cda62eabd90ffe674f296926cff202d00f78f3da1b882f09c909f4a6ef0"],"state_sha256":"96c39b9b82ce42825d2729300563d0e92fc12a4abb5ae1c8d822adb014a4eeca"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"weHqh1ye+IGp0TYwP6KX5AgyDfz1nshhjmvh/wmIrhh+vb1XpzBHSCmqTDto1WAc4YKAXTptas4mDydd/qfyAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T13:38:11.456400Z","bundle_sha256":"f7c58bf544a2004fc1b6d3e9914971ea20aecd6092f025af297fc62fd7a72dec"}}