{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:CH7J2YGBK3EYJDIWWBQYJ2ZTCB","short_pith_number":"pith:CH7J2YGB","schema_version":"1.0","canonical_sha256":"11fe9d60c156c9848d16b06184eb33107bd37d06c75b20c307bf0e178fb4725c","source":{"kind":"arxiv","id":"2312.14033","version":3},"attestation_state":"computed","paper":{"title":"T-Eval: Evaluating the Tool Utilization Capability of Large Language Models Step by Step","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dahua Lin, Feng Zhao, Jiangning Liu, Jingming Zhuo, Kai Chen, Kuikun Liu, Miao Zheng, Songyang Zhang, Weihua Du, Wenwei Zhang, Zehui Chen","submitted_at":"2023-12-21T17:02:06Z","abstract_excerpt":"Large language models (LLM) have achieved remarkable performance on various NLP tasks and are augmented by tools for broader applications. Yet, how to evaluate and analyze the tool-utilization capability of LLMs is still under-explored. In contrast to previous works that evaluate models holistically, we comprehensively decompose the tool utilization into multiple sub-processes, including instruction following, planning, reasoning, retrieval, understanding, and review. Based on that, we further introduce T-Eval to evaluate the tool utilization capability step by step. T-Eval disentangles the to"},"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":"2312.14033","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-21T17:02:06Z","cross_cats_sorted":[],"title_canon_sha256":"b825ba9fc05fdb847381138e2a12ed284250d2e855a12f7023a8958bccae65bd","abstract_canon_sha256":"391be5c0346383e3f8900d83a301da3e33987a8ffce9a7cfb11c0e2393ff01f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:33:24.717177Z","signature_b64":"GSUduK936roZQ4bVSXduu97si2zUNNu8cu+jo7G/C4hP0GZh6C3po92cVrGukwSiHIXkPfu4nxhysTKUaDSjAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"11fe9d60c156c9848d16b06184eb33107bd37d06c75b20c307bf0e178fb4725c","last_reissued_at":"2026-07-05T07:33:24.716683Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:33:24.716683Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"T-Eval: Evaluating the Tool Utilization Capability of Large Language Models Step by Step","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dahua Lin, Feng Zhao, Jiangning Liu, Jingming Zhuo, Kai Chen, Kuikun Liu, Miao Zheng, Songyang Zhang, Weihua Du, Wenwei Zhang, Zehui Chen","submitted_at":"2023-12-21T17:02:06Z","abstract_excerpt":"Large language models (LLM) have achieved remarkable performance on various NLP tasks and are augmented by tools for broader applications. Yet, how to evaluate and analyze the tool-utilization capability of LLMs is still under-explored. In contrast to previous works that evaluate models holistically, we comprehensively decompose the tool utilization into multiple sub-processes, including instruction following, planning, reasoning, retrieval, understanding, and review. Based on that, we further introduce T-Eval to evaluate the tool utilization capability step by step. T-Eval disentangles the to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.14033","kind":"arxiv","version":3},"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/2312.14033/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":"2312.14033","created_at":"2026-07-05T07:33:24.716752+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.14033v3","created_at":"2026-07-05T07:33:24.716752+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.14033","created_at":"2026-07-05T07:33:24.716752+00:00"},{"alias_kind":"pith_short_12","alias_value":"CH7J2YGBK3EY","created_at":"2026-07-05T07:33:24.716752+00:00"},{"alias_kind":"pith_short_16","alias_value":"CH7J2YGBK3EYJDIW","created_at":"2026-07-05T07:33:24.716752+00:00"},{"alias_kind":"pith_short_8","alias_value":"CH7J2YGB","created_at":"2026-07-05T07:33:24.716752+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17030","citing_title":"Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video Generation","ref_index":227,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03892","citing_title":"Synthesize and Reward -- Reinforcement Learning for Multi-Step Tool Use in Live Environments","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2403.17297","citing_title":"InternLM2 Technical Report","ref_index":193,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14865","citing_title":"Holistic Evaluation and Failure Diagnosis of AI Agents","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2507.21046","citing_title":"A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence","ref_index":93,"is_internal_anchor":false},{"citing_arxiv_id":"2507.13334","citing_title":"A Survey of Context Engineering for Large Language Models","ref_index":161,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10516","citing_title":"Consistency as a Testable Property: Statistical Methods to Evaluate AI Agent Reliability","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16752","citing_title":"Don't Start What You Can't Finish: A Counterfactual Audit of Support-State Triage in LLM Agents","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CH7J2YGBK3EYJDIWWBQYJ2ZTCB","json":"https://pith.science/pith/CH7J2YGBK3EYJDIWWBQYJ2ZTCB.json","graph_json":"https://pith.science/api/pith-number/CH7J2YGBK3EYJDIWWBQYJ2ZTCB/graph.json","events_json":"https://pith.science/api/pith-number/CH7J2YGBK3EYJDIWWBQYJ2ZTCB/events.json","paper":"https://pith.science/paper/CH7J2YGB"},"agent_actions":{"view_html":"https://pith.science/pith/CH7J2YGBK3EYJDIWWBQYJ2ZTCB","download_json":"https://pith.science/pith/CH7J2YGBK3EYJDIWWBQYJ2ZTCB.json","view_paper":"https://pith.science/paper/CH7J2YGB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.14033&json=true","fetch_graph":"https://pith.science/api/pith-number/CH7J2YGBK3EYJDIWWBQYJ2ZTCB/graph.json","fetch_events":"https://pith.science/api/pith-number/CH7J2YGBK3EYJDIWWBQYJ2ZTCB/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CH7J2YGBK3EYJDIWWBQYJ2ZTCB/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CH7J2YGBK3EYJDIWWBQYJ2ZTCB/action/storage_attestation","attest_author":"https://pith.science/pith/CH7J2YGBK3EYJDIWWBQYJ2ZTCB/action/author_attestation","sign_citation":"https://pith.science/pith/CH7J2YGBK3EYJDIWWBQYJ2ZTCB/action/citation_signature","submit_replication":"https://pith.science/pith/CH7J2YGBK3EYJDIWWBQYJ2ZTCB/action/replication_record"}},"created_at":"2026-07-05T07:33:24.716752+00:00","updated_at":"2026-07-05T07:33:24.716752+00:00"}