{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:IPDYF7WAL2JQY73ECA6M6ZKLS7","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":"9c9c6a40f471db00bc795491a595370a6b25f237c8ce7cac442ef63a1e1bb3ea","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-08T15:46:32Z","title_canon_sha256":"512f61380dab930691294793664126497ef486c232611cee5338df827f315f40"},"schema_version":"1.0","source":{"id":"2306.05301","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.05301","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"arxiv_version","alias_value":"2306.05301v2","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.05301","created_at":"2026-05-17T23:38:49Z"},{"alias_kind":"pith_short_12","alias_value":"IPDYF7WAL2JQ","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"IPDYF7WAL2JQY73E","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"IPDYF7WA","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:7517c45491ce04cf486fd3b0970be12cc91de4ce9817c650055a1834dbb2158c","target":"graph","created_at":"2026-05-17T23:38:49Z","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":"Experimental results demonstrate that ToolAlpaca achieves effective generalized tool-use capabilities comparable to those of extremely large language models like GPT-3.5, demonstrating that learning generalized tool-use ability is feasible for compact language models."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The simulated multi-agent interactions produce training data whose distribution is close enough to real-world tool use that fine-tuned models generalize to unseen APIs without additional per-tool supervision."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ToolAlpaca trains 7B and 13B models on 3938 simulated tool-use cases to reach generalized tool-use performance comparable to GPT-3.5 on unseen APIs."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Compact language models can learn to use new real-world tools by training on simulated multi-agent interactions."}],"snapshot_sha256":"c03b8e867060ecd148fde9265f234cd5d4226ebfdcfd600b990dab9c82d2d9c4"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"45df9e96c87c60b8217f373fb1c05b4802bb4e1df63fd97a244959507511895e"},"paper":{"abstract_excerpt":"Enabling large language models to utilize real-world tools effectively is crucial for achieving embodied intelligence. Existing approaches to tool learning have either primarily relied on extremely large language models, such as GPT-4, to attain generalized tool-use abilities in a zero-shot manner, or utilized supervised learning to train limited scopes of tools on compact models. However, it remains uncertain whether smaller language models can achieve generalized tool-use abilities without tool-specific training. To address this question, this paper introduces ToolAlpaca, a novel framework d","authors_text":"Boxi Cao, Hongyu Lin, Le Sun, Qiao Liang, Qiaoyu Tang, Xianpei Han, Ziliang Deng","cross_cats":[],"headline":"Compact language models can learn to use new real-world tools by training on simulated multi-agent interactions.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-08T15:46:32Z","title":"ToolAlpaca: Generalized Tool Learning for Language Models with 3000 Simulated Cases"},"references":{"count":34,"internal_anchors":2,"resolved_work":34,"sample":[{"cited_arxiv_id":"2304.08244","doi":"","is_internal_anchor":true,"ref_index":1,"title":"API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs","work_id":"a20d9332-ab34-485c-a060-1ba47cc98930","year":2023},{"cited_arxiv_id":"2303.17580","doi":"","is_internal_anchor":true,"ref_index":2,"title":"HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face","work_id":"f20ed1da-2676-4598-a11b-54549718735b","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"On the tool manipulation capability of open-source large language models.arXiv preprint arXiv:2305.16504(2023)","work_id":"18febc22-221f-4db2-8168-15e148f613c3","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Write a general overview of the API 's purpose and functionality","work_id":"4f28c4cb-f1aa-4915-97eb-fe1ba788466e","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"List and briefly describe all features provided by the API, ensuring each feature has a clear and distinct purpose with low coupling between them","work_id":"ed0e44bb-d435-423f-a67e-eb9b69f7ea8e","year":null}],"snapshot_sha256":"25d4d992dc8065cdcfcf4029f83dbf3cdb6812c2d13360a181c9ed5fed450295"},"source":{"id":"2306.05301","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-15T22:59:52.482884Z","id":"dbd30bb2-7336-4641-8cf6-00bdd5144a9e","model_set":{"reader":"grok-4.3"},"one_line_summary":"ToolAlpaca trains 7B and 13B models on 3938 simulated tool-use cases to reach generalized tool-use performance comparable to GPT-3.5 on unseen APIs.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Compact language models can learn to use new real-world tools by training on simulated multi-agent interactions.","strongest_claim":"Experimental results demonstrate that ToolAlpaca achieves effective generalized tool-use capabilities comparable to those of extremely large language models like GPT-3.5, demonstrating that learning generalized tool-use ability is feasible for compact language models.","weakest_assumption":"The simulated multi-agent interactions produce training data whose distribution is close enough to real-world tool use that fine-tuned models generalize to unseen APIs without additional per-tool supervision."}},"verdict_id":"dbd30bb2-7336-4641-8cf6-00bdd5144a9e"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:52a603ce18098926383b71a77991d6b8abb0a6c7d0f745120e03b9e27704e6f4","target":"record","created_at":"2026-05-17T23:38:49Z","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":"9c9c6a40f471db00bc795491a595370a6b25f237c8ce7cac442ef63a1e1bb3ea","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-06-08T15:46:32Z","title_canon_sha256":"512f61380dab930691294793664126497ef486c232611cee5338df827f315f40"},"schema_version":"1.0","source":{"id":"2306.05301","kind":"arxiv","version":2}},"canonical_sha256":"43c782fec05e930c7f64103ccf654b97d85382871ba7af91c76cfd7468ad415a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"43c782fec05e930c7f64103ccf654b97d85382871ba7af91c76cfd7468ad415a","first_computed_at":"2026-05-17T23:38:49.853235Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:38:49.853235Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ylXMI9O0apk2gHN9941PpGq55A/UGX2/bA3HPvTzkE60KGBSqNOkhohGX7tSZcTEdLsMEWar6VGyOZ/hZmaUAA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:38:49.853803Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.05301","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:52a603ce18098926383b71a77991d6b8abb0a6c7d0f745120e03b9e27704e6f4","sha256:7517c45491ce04cf486fd3b0970be12cc91de4ce9817c650055a1834dbb2158c"],"state_sha256":"1ec6807d394fd2e97083973e283a0dfc300cf8f409166c51f28b7138ed8b680c"}