{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:VPZHUYDC6KYKWDPZ6QRBB5BIHJ","short_pith_number":"pith:VPZHUYDC","schema_version":"1.0","canonical_sha256":"abf27a6062f2b0ab0df9f42210f4283a613c92f99cd5a18b891cf9f8e21713a7","source":{"kind":"arxiv","id":"2607.03953","version":1},"attestation_state":"computed","paper":{"title":"The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alexander Somma, Fred Premji, Isabelle Plante","submitted_at":"2026-07-04T16:59:12Z","abstract_excerpt":"This study independently replicates and extends the Natural Language Tools (NLT) framework of Johnson et al.~(2025), which questions the use of structured tool calling in large language model (LLM) agentic systems. We evaluated NLT across 14 models and 8,560 trials, adding newer frontier, reasoning, and open-weight models to the original set. The results confirm the core findings and add detail. NLT improves tool-calling accuracy by 14.9 percentage points overall (62.3\\% versus 47.4\\% structured) and reduces critical errors by 93\\% (51 versus 755 errors). The gains depend on model capability: "},"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":"2607.03953","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-04T16:59:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"e1a20436a6765594484aeadd564ef81e597de6e7139e2cc6c701505a027c58b8","abstract_canon_sha256":"42a24130f768e18e838d5d59824d8166472476ebc7c0aa4b5b38427bb1c73e31"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:18:49.417287Z","signature_b64":"L8XFZsbumBzNLmzDC6aepswKQYvOSXFm0+LkGphjJ5/prUqg6oC1fpJObOCUsprAxl/GSWT53w1j/7lRBlaCAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"abf27a6062f2b0ab0df9f42210f4283a613c92f99cd5a18b891cf9f8e21713a7","last_reissued_at":"2026-07-07T02:18:49.416347Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:18:49.416347Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Remarkable Effectiveness of Providing AI Agents with Natural Language Tools: A Replication Study Validating NLT Performance Across 14 Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alexander Somma, Fred Premji, Isabelle Plante","submitted_at":"2026-07-04T16:59:12Z","abstract_excerpt":"This study independently replicates and extends the Natural Language Tools (NLT) framework of Johnson et al.~(2025), which questions the use of structured tool calling in large language model (LLM) agentic systems. We evaluated NLT across 14 models and 8,560 trials, adding newer frontier, reasoning, and open-weight models to the original set. The results confirm the core findings and add detail. NLT improves tool-calling accuracy by 14.9 percentage points overall (62.3\\% versus 47.4\\% structured) and reduces critical errors by 93\\% (51 versus 755 errors). The gains depend on model capability: "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.03953","kind":"arxiv","version":1},"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/2607.03953/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":"2607.03953","created_at":"2026-07-07T02:18:49.416479+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.03953v1","created_at":"2026-07-07T02:18:49.416479+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.03953","created_at":"2026-07-07T02:18:49.416479+00:00"},{"alias_kind":"pith_short_12","alias_value":"VPZHUYDC6KYK","created_at":"2026-07-07T02:18:49.416479+00:00"},{"alias_kind":"pith_short_16","alias_value":"VPZHUYDC6KYKWDPZ","created_at":"2026-07-07T02:18:49.416479+00:00"},{"alias_kind":"pith_short_8","alias_value":"VPZHUYDC","created_at":"2026-07-07T02:18:49.416479+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VPZHUYDC6KYKWDPZ6QRBB5BIHJ","json":"https://pith.science/pith/VPZHUYDC6KYKWDPZ6QRBB5BIHJ.json","graph_json":"https://pith.science/api/pith-number/VPZHUYDC6KYKWDPZ6QRBB5BIHJ/graph.json","events_json":"https://pith.science/api/pith-number/VPZHUYDC6KYKWDPZ6QRBB5BIHJ/events.json","paper":"https://pith.science/paper/VPZHUYDC"},"agent_actions":{"view_html":"https://pith.science/pith/VPZHUYDC6KYKWDPZ6QRBB5BIHJ","download_json":"https://pith.science/pith/VPZHUYDC6KYKWDPZ6QRBB5BIHJ.json","view_paper":"https://pith.science/paper/VPZHUYDC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.03953&json=true","fetch_graph":"https://pith.science/api/pith-number/VPZHUYDC6KYKWDPZ6QRBB5BIHJ/graph.json","fetch_events":"https://pith.science/api/pith-number/VPZHUYDC6KYKWDPZ6QRBB5BIHJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VPZHUYDC6KYKWDPZ6QRBB5BIHJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VPZHUYDC6KYKWDPZ6QRBB5BIHJ/action/storage_attestation","attest_author":"https://pith.science/pith/VPZHUYDC6KYKWDPZ6QRBB5BIHJ/action/author_attestation","sign_citation":"https://pith.science/pith/VPZHUYDC6KYKWDPZ6QRBB5BIHJ/action/citation_signature","submit_replication":"https://pith.science/pith/VPZHUYDC6KYKWDPZ6QRBB5BIHJ/action/replication_record"}},"created_at":"2026-07-07T02:18:49.416479+00:00","updated_at":"2026-07-07T02:18:49.416479+00:00"}