{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:RD6JRUX4SMR3NC3MVFS4EZZMVH","short_pith_number":"pith:RD6JRUX4","schema_version":"1.0","canonical_sha256":"88fc98d2fc9323b68b6ca965c2672ca9c4b9ec2f4105c1cbe4221274a8fe0827","source":{"kind":"arxiv","id":"2210.09150","version":2},"attestation_state":"computed","paper":{"title":"Prompting GPT-3 To Be Reliable","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenglei Si, Jianfeng Wang, Jordan Boyd-Graber, Lijuan Wang, Shuohang Wang, Zhe Gan, Zhengyuan Yang","submitted_at":"2022-10-17T14:52:39Z","abstract_excerpt":"Large language models (LLMs) show impressive abilities via few-shot prompting. Commercialized APIs such as OpenAI GPT-3 further increase their use in real-world language applications. However, the crucial problem of how to improve the reliability of GPT-3 is still under-explored. While reliability is a broad and vaguely defined term, we decompose reliability into four main facets that correspond to the existing framework of ML safety and are well-recognized to be important: generalizability, social biases, calibration, and factuality. Our core contribution is to establish simple and effective "},"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":"2210.09150","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-10-17T14:52:39Z","cross_cats_sorted":[],"title_canon_sha256":"8e0ee1087be6ed2ab708c7bd1475f161b4b94cad82034e1ac0557df3e54179f8","abstract_canon_sha256":"b3a1e8845bfbc760d2b054b421a5d76bfb4796bf136f622f1520704d1f9b8ded"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:41:51.482753Z","signature_b64":"V/fpQSWQ/axci3Hemc4MX4R/jpWMQW7zy81BX0sBbsmVSD6r3baEnduHU9gPggggegshb2zYN4quyAGvjsPVDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88fc98d2fc9323b68b6ca965c2672ca9c4b9ec2f4105c1cbe4221274a8fe0827","last_reissued_at":"2026-07-05T05:41:51.482265Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:41:51.482265Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Prompting GPT-3 To Be Reliable","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chenglei Si, Jianfeng Wang, Jordan Boyd-Graber, Lijuan Wang, Shuohang Wang, Zhe Gan, Zhengyuan Yang","submitted_at":"2022-10-17T14:52:39Z","abstract_excerpt":"Large language models (LLMs) show impressive abilities via few-shot prompting. Commercialized APIs such as OpenAI GPT-3 further increase their use in real-world language applications. However, the crucial problem of how to improve the reliability of GPT-3 is still under-explored. While reliability is a broad and vaguely defined term, we decompose reliability into four main facets that correspond to the existing framework of ML safety and are well-recognized to be important: generalizability, social biases, calibration, and factuality. Our core contribution is to establish simple and effective "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.09150","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/2210.09150/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":"2210.09150","created_at":"2026-07-05T05:41:51.482327+00:00"},{"alias_kind":"arxiv_version","alias_value":"2210.09150v2","created_at":"2026-07-05T05:41:51.482327+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.09150","created_at":"2026-07-05T05:41:51.482327+00:00"},{"alias_kind":"pith_short_12","alias_value":"RD6JRUX4SMR3","created_at":"2026-07-05T05:41:51.482327+00:00"},{"alias_kind":"pith_short_16","alias_value":"RD6JRUX4SMR3NC3M","created_at":"2026-07-05T05:41:51.482327+00:00"},{"alias_kind":"pith_short_8","alias_value":"RD6JRUX4","created_at":"2026-07-05T05:41:51.482327+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03535","citing_title":"Can LLM Rerankers Predict Their Own Ranking Performance?","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2307.06435","citing_title":"A Comprehensive Overview of Large Language Models","ref_index":175,"is_internal_anchor":false},{"citing_arxiv_id":"2308.05374","citing_title":"Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignment","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2301.12652","citing_title":"REPLUG: Retrieval-Augmented Black-Box Language Models","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2401.01313","citing_title":"A Comprehensive Survey of Hallucination Mitigation Techniques in Large Language Models","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2310.03714","citing_title":"DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines","ref_index":51,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15741","citing_title":"Learning Uncertainty from Sequential Internal Dispersion in Large Language Models","ref_index":42,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RD6JRUX4SMR3NC3MVFS4EZZMVH","json":"https://pith.science/pith/RD6JRUX4SMR3NC3MVFS4EZZMVH.json","graph_json":"https://pith.science/api/pith-number/RD6JRUX4SMR3NC3MVFS4EZZMVH/graph.json","events_json":"https://pith.science/api/pith-number/RD6JRUX4SMR3NC3MVFS4EZZMVH/events.json","paper":"https://pith.science/paper/RD6JRUX4"},"agent_actions":{"view_html":"https://pith.science/pith/RD6JRUX4SMR3NC3MVFS4EZZMVH","download_json":"https://pith.science/pith/RD6JRUX4SMR3NC3MVFS4EZZMVH.json","view_paper":"https://pith.science/paper/RD6JRUX4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2210.09150&json=true","fetch_graph":"https://pith.science/api/pith-number/RD6JRUX4SMR3NC3MVFS4EZZMVH/graph.json","fetch_events":"https://pith.science/api/pith-number/RD6JRUX4SMR3NC3MVFS4EZZMVH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RD6JRUX4SMR3NC3MVFS4EZZMVH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RD6JRUX4SMR3NC3MVFS4EZZMVH/action/storage_attestation","attest_author":"https://pith.science/pith/RD6JRUX4SMR3NC3MVFS4EZZMVH/action/author_attestation","sign_citation":"https://pith.science/pith/RD6JRUX4SMR3NC3MVFS4EZZMVH/action/citation_signature","submit_replication":"https://pith.science/pith/RD6JRUX4SMR3NC3MVFS4EZZMVH/action/replication_record"}},"created_at":"2026-07-05T05:41:51.482327+00:00","updated_at":"2026-07-05T05:41:51.482327+00:00"}