{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:BAC4OOTTCJLZL4ETBDI6ZCW56B","short_pith_number":"pith:BAC4OOTT","schema_version":"1.0","canonical_sha256":"0805c73a73125795f09308d1ec8addf073d5269064c9a5caf3e38ce77a70e2c7","source":{"kind":"arxiv","id":"2306.00622","version":1},"attestation_state":"computed","paper":{"title":"ReviewerGPT? An Exploratory Study on Using Large Language Models for Paper Reviewing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.DL"],"primary_cat":"cs.CL","authors_text":"Nihar B. Shah, Ryan Liu","submitted_at":"2023-06-01T12:45:53Z","abstract_excerpt":"Given the rapid ascent of large language models (LLMs), we study the question: (How) can large language models help in reviewing of scientific papers or proposals? We first conduct some pilot studies where we find that (i) GPT-4 outperforms other LLMs (Bard, Vicuna, Koala, Alpaca, LLaMa, Dolly, OpenAssistant, StableLM), and (ii) prompting with a specific question (e.g., to identify errors) outperforms prompting to simply write a review. With these insights, we study the use of LLMs (specifically, GPT-4) for three tasks:\n  1. Identifying errors: We construct 13 short computer science papers eac"},"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":"2306.00622","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-06-01T12:45:53Z","cross_cats_sorted":["cs.AI","cs.DL"],"title_canon_sha256":"5e7c7b23e294338bfbd2867708156fbd43f6f5ef778d70cba17be8f4783ce17c","abstract_canon_sha256":"df52691a77b7a62b1ddea18267ae54d1c44aaff25f33681904a2c85a0f28edf7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:16:28.874043Z","signature_b64":"+nT7nm4EhRlVtEHapDTCCQxuW8vEyvNXO4mHrsFztnn/ewq5Vg8+bOJ4j8hAM+Xlpwhkd6sPI9q3hh4QVfgqDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0805c73a73125795f09308d1ec8addf073d5269064c9a5caf3e38ce77a70e2c7","last_reissued_at":"2026-07-05T06:16:28.873575Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:16:28.873575Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ReviewerGPT? An Exploratory Study on Using Large Language Models for Paper Reviewing","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.DL"],"primary_cat":"cs.CL","authors_text":"Nihar B. Shah, Ryan Liu","submitted_at":"2023-06-01T12:45:53Z","abstract_excerpt":"Given the rapid ascent of large language models (LLMs), we study the question: (How) can large language models help in reviewing of scientific papers or proposals? We first conduct some pilot studies where we find that (i) GPT-4 outperforms other LLMs (Bard, Vicuna, Koala, Alpaca, LLaMa, Dolly, OpenAssistant, StableLM), and (ii) prompting with a specific question (e.g., to identify errors) outperforms prompting to simply write a review. With these insights, we study the use of LLMs (specifically, GPT-4) for three tasks:\n  1. Identifying errors: We construct 13 short computer science papers eac"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.00622","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/2306.00622/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":"2306.00622","created_at":"2026-07-05T06:16:28.873629+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.00622v1","created_at":"2026-07-05T06:16:28.873629+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.00622","created_at":"2026-07-05T06:16:28.873629+00:00"},{"alias_kind":"pith_short_12","alias_value":"BAC4OOTTCJLZ","created_at":"2026-07-05T06:16:28.873629+00:00"},{"alias_kind":"pith_short_16","alias_value":"BAC4OOTTCJLZL4ET","created_at":"2026-07-05T06:16:28.873629+00:00"},{"alias_kind":"pith_short_8","alias_value":"BAC4OOTT","created_at":"2026-07-05T06:16:28.873629+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":14,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24530","citing_title":"NatureBench: Can Coding Agents Match the Published SOTA of Nature-Family Papers?","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21228","citing_title":"Sakana Fugu Technical Report","ref_index":151,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01240","citing_title":"Prompt Framing Distorts Count Based Evaluation of LLM Error Detection: Evidence from Numeric Anchoring","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18237","citing_title":"ReproRepo: Scaling Reproducibility Audits with GitHub Repository Issues","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08857","citing_title":"PaperMentor: A Human-Centered Multi-Agent Writing Tutor for AI Research Papers on Overleaf","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2606.13349","citing_title":"From Passive Generation to Investigation: A Proactive Scientific Peer Review Agent","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2403.07183","citing_title":"Monitoring AI-Modified Content at Scale: A Case Study on the Impact of ChatGPT on AI Conference Peer Reviews","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07855","citing_title":"Jagged AI in Scientific Peer Review: Evidence from POMP Data Analysis","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2601.03013","citing_title":"LLMs, You Can Evaluate It! Design of Multi-perspective Report Evaluation for Security Operation Centers","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2509.19349","citing_title":"ShinkaEvolve: Towards Open-Ended And Sample-Efficient Program Evolution","ref_index":115,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19578","citing_title":"Impact of large language models on peer review opinions from a fine-grained perspective: Evidence from top conference proceedings in AI","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07855","citing_title":"Jagged AI in Scientific Peer Review: Evidence from POMP Data Analysis","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13940","citing_title":"AI-Assisted Peer Review at Scale: The AAAI-26 AI Review Pilot","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03042","citing_title":"ARIS: Autonomous Research via Adversarial Multi-Agent Collaboration","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BAC4OOTTCJLZL4ETBDI6ZCW56B","json":"https://pith.science/pith/BAC4OOTTCJLZL4ETBDI6ZCW56B.json","graph_json":"https://pith.science/api/pith-number/BAC4OOTTCJLZL4ETBDI6ZCW56B/graph.json","events_json":"https://pith.science/api/pith-number/BAC4OOTTCJLZL4ETBDI6ZCW56B/events.json","paper":"https://pith.science/paper/BAC4OOTT"},"agent_actions":{"view_html":"https://pith.science/pith/BAC4OOTTCJLZL4ETBDI6ZCW56B","download_json":"https://pith.science/pith/BAC4OOTTCJLZL4ETBDI6ZCW56B.json","view_paper":"https://pith.science/paper/BAC4OOTT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.00622&json=true","fetch_graph":"https://pith.science/api/pith-number/BAC4OOTTCJLZL4ETBDI6ZCW56B/graph.json","fetch_events":"https://pith.science/api/pith-number/BAC4OOTTCJLZL4ETBDI6ZCW56B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BAC4OOTTCJLZL4ETBDI6ZCW56B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BAC4OOTTCJLZL4ETBDI6ZCW56B/action/storage_attestation","attest_author":"https://pith.science/pith/BAC4OOTTCJLZL4ETBDI6ZCW56B/action/author_attestation","sign_citation":"https://pith.science/pith/BAC4OOTTCJLZL4ETBDI6ZCW56B/action/citation_signature","submit_replication":"https://pith.science/pith/BAC4OOTTCJLZL4ETBDI6ZCW56B/action/replication_record"}},"created_at":"2026-07-05T06:16:28.873629+00:00","updated_at":"2026-07-05T06:16:28.873629+00:00"}