{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:36NPVMZXJERDM3WOZPFW4COSWV","short_pith_number":"pith:36NPVMZX","schema_version":"1.0","canonical_sha256":"df9afab3374922366ececbcb6e09d2b56027d503db0bc424ee63c69f35af2724","source":{"kind":"arxiv","id":"2508.20931","version":2},"attestation_state":"computed","paper":{"title":"How Can Input Reformulation Improve Tool Usage Accuracy in a Complex Dynamic Environment? A Study on $\\tau$-bench","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ali Payani, Amir Saeidi, Chitta Baral, Gaowen Liu, Jayanth Srinivasa, Mutsumi Nakamura, Satyam Raj, Venkatesh Mishra","submitted_at":"2025-08-28T15:57:33Z","abstract_excerpt":"Recent advances in reasoning and planning capabilities of large language models (LLMs) have enabled their potential as autonomous agents capable of tool use in dynamic environments. However, in multi-turn conversational environments like $\\tau$-bench, these agents often struggle with consistent reasoning, adherence to domain-specific policies, and extracting correct information over a long horizon of tool-calls and conversation. To capture and mitigate these failures, we conduct a comprehensive manual analysis of the common errors occurring in the conversation trajectories. We then experiment "},"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":"2508.20931","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-28T15:57:33Z","cross_cats_sorted":[],"title_canon_sha256":"b6217394cf02009a4b92a8475e62317f142d80133de6413e047e5593496bb4b2","abstract_canon_sha256":"9dae32430fcec83cf184f379d421588cea4636c3553b95ac6fae9d637f343ea9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:02:51.684924Z","signature_b64":"J905mRvJfdNX7amX2r62fSJUJtIw0Jb46MUzJLciW1ZNr7yNd6jNZyyOdOCe+/cMPVkXV+LDYREzkAm10HdIBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"df9afab3374922366ececbcb6e09d2b56027d503db0bc424ee63c69f35af2724","last_reissued_at":"2026-07-05T12:02:51.684436Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:02:51.684436Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How Can Input Reformulation Improve Tool Usage Accuracy in a Complex Dynamic Environment? A Study on $\\tau$-bench","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ali Payani, Amir Saeidi, Chitta Baral, Gaowen Liu, Jayanth Srinivasa, Mutsumi Nakamura, Satyam Raj, Venkatesh Mishra","submitted_at":"2025-08-28T15:57:33Z","abstract_excerpt":"Recent advances in reasoning and planning capabilities of large language models (LLMs) have enabled their potential as autonomous agents capable of tool use in dynamic environments. However, in multi-turn conversational environments like $\\tau$-bench, these agents often struggle with consistent reasoning, adherence to domain-specific policies, and extracting correct information over a long horizon of tool-calls and conversation. To capture and mitigate these failures, we conduct a comprehensive manual analysis of the common errors occurring in the conversation trajectories. We then experiment "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.20931","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/2508.20931/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":"2508.20931","created_at":"2026-07-05T12:02:51.684485+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.20931v2","created_at":"2026-07-05T12:02:51.684485+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.20931","created_at":"2026-07-05T12:02:51.684485+00:00"},{"alias_kind":"pith_short_12","alias_value":"36NPVMZXJERD","created_at":"2026-07-05T12:02:51.684485+00:00"},{"alias_kind":"pith_short_16","alias_value":"36NPVMZXJERDM3WO","created_at":"2026-07-05T12:02:51.684485+00:00"},{"alias_kind":"pith_short_8","alias_value":"36NPVMZX","created_at":"2026-07-05T12:02:51.684485+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23112","citing_title":"Self-Evolution for Multi-Turn Tool-Calling Agents via Divergence-Point Preference Learning","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/36NPVMZXJERDM3WOZPFW4COSWV","json":"https://pith.science/pith/36NPVMZXJERDM3WOZPFW4COSWV.json","graph_json":"https://pith.science/api/pith-number/36NPVMZXJERDM3WOZPFW4COSWV/graph.json","events_json":"https://pith.science/api/pith-number/36NPVMZXJERDM3WOZPFW4COSWV/events.json","paper":"https://pith.science/paper/36NPVMZX"},"agent_actions":{"view_html":"https://pith.science/pith/36NPVMZXJERDM3WOZPFW4COSWV","download_json":"https://pith.science/pith/36NPVMZXJERDM3WOZPFW4COSWV.json","view_paper":"https://pith.science/paper/36NPVMZX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.20931&json=true","fetch_graph":"https://pith.science/api/pith-number/36NPVMZXJERDM3WOZPFW4COSWV/graph.json","fetch_events":"https://pith.science/api/pith-number/36NPVMZXJERDM3WOZPFW4COSWV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/36NPVMZXJERDM3WOZPFW4COSWV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/36NPVMZXJERDM3WOZPFW4COSWV/action/storage_attestation","attest_author":"https://pith.science/pith/36NPVMZXJERDM3WOZPFW4COSWV/action/author_attestation","sign_citation":"https://pith.science/pith/36NPVMZXJERDM3WOZPFW4COSWV/action/citation_signature","submit_replication":"https://pith.science/pith/36NPVMZXJERDM3WOZPFW4COSWV/action/replication_record"}},"created_at":"2026-07-05T12:02:51.684485+00:00","updated_at":"2026-07-05T12:02:51.684485+00:00"}