{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:EJVCKAVCH6TQ4G5ONYV5RXEWIT","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":"fc18cfe94626e98794805db81b978304805750ed8622ad370687008a6b807ec8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-03-16T01:04:45Z","title_canon_sha256":"cad4a37e49b9b66ccfa0ac42edf9a2ea6626c8aa4df881435177bfde702d5028"},"schema_version":"1.0","source":{"id":"2303.09014","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2303.09014","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"arxiv_version","alias_value":"2303.09014v1","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2303.09014","created_at":"2026-05-17T23:38:46Z"},{"alias_kind":"pith_short_12","alias_value":"EJVCKAVCH6TQ","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_16","alias_value":"EJVCKAVCH6TQ4G5O","created_at":"2026-05-18T12:33:33Z"},{"alias_kind":"pith_short_8","alias_value":"EJVCKAVC","created_at":"2026-05-18T12:33:33Z"}],"graph_snapshots":[{"event_id":"sha256:bd8241b160331094242f4fb26b325667256eec52fa5966b7cb2afca49e6a4dbb","target":"graph","created_at":"2026-05-17T23:38:46Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"ART achieves a substantial improvement over few-shot prompting and automatic CoT on unseen tasks in the BigBench and MMLU benchmarks, and matches performance of hand-crafted CoT prompts on a majority of these tasks."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That a fixed task library contains sufficiently diverse and high-quality demonstrations so that nearest-neighbor selection reliably supplies useful programs for entirely new, unseen tasks."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ART automatically generates multi-step reasoning programs with tool integration for LLMs, yielding substantial gains over few-shot and auto-CoT prompting on BigBench and MMLU while matching hand-crafted CoT on most tasks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"ART lets large language models automatically generate multi-step reasoning programs that call external tools."}],"snapshot_sha256":"28d7690fa1db8d61d0b9b4bc61bdeadc255b91c797f197c58f2bfa9570b91f50"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"8066222b8c40f1850eb92d5a7d0a32d7eeb36653bd62ca4232f5f123348f7d8e"},"paper":{"abstract_excerpt":"Large language models (LLMs) can perform complex reasoning in few- and zero-shot settings by generating intermediate chain of thought (CoT) reasoning steps. Further, each reasoning step can rely on external tools to support computation beyond the core LLM capabilities (e.g. search/running code). Prior work on CoT prompting and tool use typically requires hand-crafting task-specific demonstrations and carefully scripted interleaving of model generations with tool use. We introduce Automatic Reasoning and Tool-use (ART), a framework that uses frozen LLMs to automatically generate intermediate re","authors_text":"Bhargavi Paranjape, Hannaneh Hajishirzi, Luke Zettlemoyer, Marco Tulio Ribeiro, Sameer Singh, Scott Lundberg","cross_cats":[],"headline":"ART lets large language models automatically generate multi-step reasoning programs that call external tools.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-03-16T01:04:45Z","title":"ART: Automatic multi-step reasoning and tool-use for large language models"},"references":{"count":267,"internal_anchors":38,"resolved_work":267,"sample":[{"cited_arxiv_id":"","doi":"10.18653/v1/2022.acl-long.579","is_internal_anchor":false,"ref_index":15,"title":"Mojtaba Komeili, Kurt Shuster, and Jason Weston. 2022. https://doi.org/10.18653/v1/2022.acl-long.579 I nternet-augmented dialogue generation . In Proceedings of the 60th Annual Meeting of the Associat","work_id":"4b63a530-9990-4f21-a11b-29062ce01cc7","year":2022},{"cited_arxiv_id":"2103.07191","doi":"10.18653/v1/2021.naacl-main.168","is_internal_anchor":true,"ref_index":22,"title":"Are NLP Models really able to Solve Simple Math Word Problems?","work_id":"01c9a44a-5ebf-4209-a628-5f1fd19d2103","year":2021},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":33,"title":"Chain-of-thought prompting elicits reasoning in large language models","work_id":"f4edf7fa-4b0c-45e2-86ab-96e1d16f24bb","year":null},{"cited_arxiv_id":"2109.01652","doi":"","is_internal_anchor":true,"ref_index":37,"title":"Finetuned Language Models Are Zero-Shot Learners","work_id":"7ed6cdaa-ed67-4db4-aceb-b7e1b0e6e7c4","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":38,"title":"Chain of Thought Prompting Elicits Reasoning in Large Language Models , author=. ArXiv , year=","work_id":"75eacb63-2829-4757-8923-44fbe8ccdd56","year":null}],"snapshot_sha256":"da9fa28cbe2d3794c22ddf8a4347b1d0c40e7c8f03a53f49ec8b01fd2f9f98f7"},"source":{"id":"2303.09014","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-16T18:58:48.398414Z","id":"ad0922ee-78ed-4279-a5e7-512d55e1bcfc","model_set":{"reader":"grok-4.3"},"one_line_summary":"ART automatically generates multi-step reasoning programs with tool integration for LLMs, yielding substantial gains over few-shot and auto-CoT prompting on BigBench and MMLU while matching hand-crafted CoT on most tasks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"ART lets large language models automatically generate multi-step reasoning programs that call external tools.","strongest_claim":"ART achieves a substantial improvement over few-shot prompting and automatic CoT on unseen tasks in the BigBench and MMLU benchmarks, and matches performance of hand-crafted CoT prompts on a majority of these tasks.","weakest_assumption":"That a fixed task library contains sufficiently diverse and high-quality demonstrations so that nearest-neighbor selection reliably supplies useful programs for entirely new, unseen tasks."}},"verdict_id":"ad0922ee-78ed-4279-a5e7-512d55e1bcfc"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:e5e5a851fee895f4bed5439ecc07885009ec5e567e09b2e3c6c75abfc92c2258","target":"record","created_at":"2026-05-17T23:38:46Z","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":"fc18cfe94626e98794805db81b978304805750ed8622ad370687008a6b807ec8","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-03-16T01:04:45Z","title_canon_sha256":"cad4a37e49b9b66ccfa0ac42edf9a2ea6626c8aa4df881435177bfde702d5028"},"schema_version":"1.0","source":{"id":"2303.09014","kind":"arxiv","version":1}},"canonical_sha256":"226a2502a23fa70e1bae6e2bd8dc9644d215cffc162551cfe5ac94e7997a9c91","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"226a2502a23fa70e1bae6e2bd8dc9644d215cffc162551cfe5ac94e7997a9c91","first_computed_at":"2026-05-17T23:38:46.903881Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:46.903881Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"X8avq66rFKLjpQARsbbZlU06iGST1XkzkcNEFn0U540oZ7BYhAktLHT/DzV1frYEUylKYk7xmDOHqfE0dWI6DQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:46.904457Z","signed_message":"canonical_sha256_bytes"},"source_id":"2303.09014","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e5e5a851fee895f4bed5439ecc07885009ec5e567e09b2e3c6c75abfc92c2258","sha256:bd8241b160331094242f4fb26b325667256eec52fa5966b7cb2afca49e6a4dbb"],"state_sha256":"758cfa047b6ad0ae7604b1dcf7247db0b3cc7f7e2e6dc6365e3aa22dd1bcc4d3"}