{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FKDMI3QL7IL2RRHL6PF6V635IK","short_pith_number":"pith:FKDMI3QL","schema_version":"1.0","canonical_sha256":"2a86c46e0bfa17a8c4ebf3cbeafb7d428b6bcee441804a83256b723d86dafb04","source":{"kind":"arxiv","id":"2406.20015","version":2},"attestation_state":"computed","paper":{"title":"ToolBeHonest: A Multi-level Hallucination Diagnostic Benchmark for Tool-Augmented Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cheng Yang, Chufan Shi, Hanwen Wan, Hayato Yamana, Jing Chen, Junjie Wang, Tetsuya Sakai, Tian Feng, Xinyu Zhu, Yaxin Liu, Yujiu Yang, Yuxiang Zhang, Zihao Lin","submitted_at":"2024-06-28T16:03:30Z","abstract_excerpt":"Tool-augmented large language models (LLMs) are rapidly being integrated into real-world applications. Due to the lack of benchmarks, the community has yet to fully understand the hallucination issues within these models. To address this challenge, we introduce a comprehensive diagnostic benchmark, ToolBH. Specifically, we assess the LLM's hallucinations through two perspectives: depth and breadth. In terms of depth, we propose a multi-level diagnostic process, including (1) solvability detection, (2) solution planning, and (3) missing-tool analysis. For breadth, we consider three scenarios ba"},"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":"2406.20015","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-28T16:03:30Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ae0c461e4733af3e490e00e6f7e533b80a0ca71ca221946a262a59a99735fb72","abstract_canon_sha256":"a4056fd9bd3939be6462a1ea31cc57e0bf91152a9dd7e991faa04f61136db3bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:34.505254Z","signature_b64":"ilFWQgvYEWqfuthXiMQ4eVmi08CYMqVzJN1xUrGSvhhqPtRkVYQOqdcSyV7s7CCzjs2D/JKOVtRW5AgWFkCQDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2a86c46e0bfa17a8c4ebf3cbeafb7d428b6bcee441804a83256b723d86dafb04","last_reissued_at":"2026-07-05T09:15:34.504686Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:34.504686Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ToolBeHonest: A Multi-level Hallucination Diagnostic Benchmark for Tool-Augmented Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Cheng Yang, Chufan Shi, Hanwen Wan, Hayato Yamana, Jing Chen, Junjie Wang, Tetsuya Sakai, Tian Feng, Xinyu Zhu, Yaxin Liu, Yujiu Yang, Yuxiang Zhang, Zihao Lin","submitted_at":"2024-06-28T16:03:30Z","abstract_excerpt":"Tool-augmented large language models (LLMs) are rapidly being integrated into real-world applications. Due to the lack of benchmarks, the community has yet to fully understand the hallucination issues within these models. To address this challenge, we introduce a comprehensive diagnostic benchmark, ToolBH. Specifically, we assess the LLM's hallucinations through two perspectives: depth and breadth. In terms of depth, we propose a multi-level diagnostic process, including (1) solvability detection, (2) solution planning, and (3) missing-tool analysis. For breadth, we consider three scenarios ba"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.20015","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/2406.20015/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":"2406.20015","created_at":"2026-07-05T09:15:34.504762+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.20015v2","created_at":"2026-07-05T09:15:34.504762+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.20015","created_at":"2026-07-05T09:15:34.504762+00:00"},{"alias_kind":"pith_short_12","alias_value":"FKDMI3QL7IL2","created_at":"2026-07-05T09:15:34.504762+00:00"},{"alias_kind":"pith_short_16","alias_value":"FKDMI3QL7IL2RRHL","created_at":"2026-07-05T09:15:34.504762+00:00"},{"alias_kind":"pith_short_8","alias_value":"FKDMI3QL","created_at":"2026-07-05T09:15:34.504762+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.27209","citing_title":"Learning to Act under Noise: Enhancing Agent Robustness via Noisy Environments","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14038","citing_title":"Model-Adaptive Tool Necessity Reveals the Knowing-Doing Gap in LLM Tool Use","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2510.02837","citing_title":"Beyond the Final Answer: Evaluating the Reasoning Trajectories of Tool-Augmented Agents","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2510.23853","citing_title":"Your LLM Agents are Temporally Blind: The Misalignment Between Tool Use Decisions and Human Time Perception","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14038","citing_title":"Model-Adaptive Tool Necessity Reveals the Knowing-Doing Gap in LLM Tool Use","ref_index":36,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FKDMI3QL7IL2RRHL6PF6V635IK","json":"https://pith.science/pith/FKDMI3QL7IL2RRHL6PF6V635IK.json","graph_json":"https://pith.science/api/pith-number/FKDMI3QL7IL2RRHL6PF6V635IK/graph.json","events_json":"https://pith.science/api/pith-number/FKDMI3QL7IL2RRHL6PF6V635IK/events.json","paper":"https://pith.science/paper/FKDMI3QL"},"agent_actions":{"view_html":"https://pith.science/pith/FKDMI3QL7IL2RRHL6PF6V635IK","download_json":"https://pith.science/pith/FKDMI3QL7IL2RRHL6PF6V635IK.json","view_paper":"https://pith.science/paper/FKDMI3QL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.20015&json=true","fetch_graph":"https://pith.science/api/pith-number/FKDMI3QL7IL2RRHL6PF6V635IK/graph.json","fetch_events":"https://pith.science/api/pith-number/FKDMI3QL7IL2RRHL6PF6V635IK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FKDMI3QL7IL2RRHL6PF6V635IK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FKDMI3QL7IL2RRHL6PF6V635IK/action/storage_attestation","attest_author":"https://pith.science/pith/FKDMI3QL7IL2RRHL6PF6V635IK/action/author_attestation","sign_citation":"https://pith.science/pith/FKDMI3QL7IL2RRHL6PF6V635IK/action/citation_signature","submit_replication":"https://pith.science/pith/FKDMI3QL7IL2RRHL6PF6V635IK/action/replication_record"}},"created_at":"2026-07-05T09:15:34.504762+00:00","updated_at":"2026-07-05T09:15:34.504762+00:00"}