{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:BQNNVBBJ34AFQVJ6U75OZELHZJ","short_pith_number":"pith:BQNNVBBJ","schema_version":"1.0","canonical_sha256":"0c1ada8429df0058553ea7faec9167ca61b65e1916dfdda3830ff08938747085","source":{"kind":"arxiv","id":"2605.09252","version":2},"attestation_state":"computed","paper":{"title":"LLM Agents Already Know When to Call Tools -- Even Without Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"LLM agents already encode whether a tool is needed in their hidden states before generating any output.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chung-En Sun, Ge Yan, Linbo Liu, Tsui-Wei Weng, Zimo Wang","submitted_at":"2026-05-10T01:37:40Z","abstract_excerpt":"Tool-augmented LLM agents tend to call tools indiscriminately, even when the model can answer directly. Each unnecessary call wastes API fees and latency, yet no existing benchmark systematically studies when a tool call is actually needed. We propose When2Tool, a benchmark of 18 environments (15 single-hop, 3 multi-hop) spanning three categories of tool necessity -- computational scale, knowledge boundaries, and execution reliability -- each with controlled difficulty levels that create a clear decision boundary between tool-necessary and tool-unnecessary tasks. We evaluate two families of tr"},"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":true},"canonical_record":{"source":{"id":"2605.09252","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-10T01:37:40Z","cross_cats_sorted":[],"title_canon_sha256":"c8e106fb54f734422ae3421a324836d082fe177bb5f4288476ca9d9e935ce317","abstract_canon_sha256":"99cf9536e1412fc5ddbb1e5237a7790d01959c8a3d3582c5cbc580429ec1a4a7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:03:20.038764Z","signature_b64":"9sG8xjNt31JkWk6BQfHJ8jSykIE1bbmOpBSUsSZL2msYKbPqCs+EwsIzoDxDW4hFvR4vRFiflFhC9en6kfcnDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c1ada8429df0058553ea7faec9167ca61b65e1916dfdda3830ff08938747085","last_reissued_at":"2026-05-22T01:03:20.038088Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:03:20.038088Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLM Agents Already Know When to Call Tools -- Even Without Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"LLM agents already encode whether a tool is needed in their hidden states before generating any output.","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chung-En Sun, Ge Yan, Linbo Liu, Tsui-Wei Weng, Zimo Wang","submitted_at":"2026-05-10T01:37:40Z","abstract_excerpt":"Tool-augmented LLM agents tend to call tools indiscriminately, even when the model can answer directly. Each unnecessary call wastes API fees and latency, yet no existing benchmark systematically studies when a tool call is actually needed. We propose When2Tool, a benchmark of 18 environments (15 single-hop, 3 multi-hop) spanning three categories of tool necessity -- computational scale, knowledge boundaries, and execution reliability -- each with controlled difficulty levels that create a clear decision boundary between tool-necessary and tool-unnecessary tasks. We evaluate two families of tr"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"tool necessity is linearly decodable from the pre-generation representation with AUROC 0.89--0.96 across six models, substantially exceeding the model's own verbalized reasoning.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The linear probe trained on hidden states from the When2Tool benchmark will generalize to new tasks and models without substantial degradation or task-specific artifacts.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LLMs encode tool necessity in pre-generation hidden states at AUROC 0.89-0.96, enabling Probe&Prefill to reduce tool calls 48% with 1.7% accuracy loss, outperforming prompt and reasoning baselines.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"LLM agents already encode whether a tool is needed in their hidden states before generating any output.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"790595cfe58e92e1c16bab1597804a24d4faf98df0ff9cc03a92820cf7ab7ec1"},"source":{"id":"2605.09252","kind":"arxiv","version":2},"verdict":{"id":"b3a95362-a864-48cf-aafa-22af8d956278","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T02:37:05.851364Z","strongest_claim":"tool necessity is linearly decodable from the pre-generation representation with AUROC 0.89--0.96 across six models, substantially exceeding the model's own verbalized reasoning.","one_line_summary":"LLMs encode tool necessity in pre-generation hidden states at AUROC 0.89-0.96, enabling Probe&Prefill to reduce tool calls 48% with 1.7% accuracy loss, outperforming prompt and reasoning baselines.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The linear probe trained on hidden states from the When2Tool benchmark will generalize to new tasks and models without substantial degradation or task-specific artifacts.","pith_extraction_headline":"LLM agents already encode whether a tool is needed in their hidden states before generating any output."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.09252/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T08:02:09.172589Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T20:34:43.494698Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T13:31:17.958794Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T10:22:49.526877Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"bb6d31f5a88a142fcb03c12860ccb3f5bda74f798277f0b4cea0675a0655eb38"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"77e33f81b3256ad82f36fe162221af29dfe2357269e3eaa2b9a73da7879b7424"},"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":"2605.09252","created_at":"2026-05-22T01:03:20.038184+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.09252v2","created_at":"2026-05-22T01:03:20.038184+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.09252","created_at":"2026-05-22T01:03:20.038184+00:00"},{"alias_kind":"pith_short_12","alias_value":"BQNNVBBJ34AF","created_at":"2026-05-22T01:03:20.038184+00:00"},{"alias_kind":"pith_short_16","alias_value":"BQNNVBBJ34AFQVJ6","created_at":"2026-05-22T01:03:20.038184+00:00"},{"alias_kind":"pith_short_8","alias_value":"BQNNVBBJ","created_at":"2026-05-22T01:03:20.038184+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BQNNVBBJ34AFQVJ6U75OZELHZJ","json":"https://pith.science/pith/BQNNVBBJ34AFQVJ6U75OZELHZJ.json","graph_json":"https://pith.science/api/pith-number/BQNNVBBJ34AFQVJ6U75OZELHZJ/graph.json","events_json":"https://pith.science/api/pith-number/BQNNVBBJ34AFQVJ6U75OZELHZJ/events.json","paper":"https://pith.science/paper/BQNNVBBJ"},"agent_actions":{"view_html":"https://pith.science/pith/BQNNVBBJ34AFQVJ6U75OZELHZJ","download_json":"https://pith.science/pith/BQNNVBBJ34AFQVJ6U75OZELHZJ.json","view_paper":"https://pith.science/paper/BQNNVBBJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.09252&json=true","fetch_graph":"https://pith.science/api/pith-number/BQNNVBBJ34AFQVJ6U75OZELHZJ/graph.json","fetch_events":"https://pith.science/api/pith-number/BQNNVBBJ34AFQVJ6U75OZELHZJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BQNNVBBJ34AFQVJ6U75OZELHZJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BQNNVBBJ34AFQVJ6U75OZELHZJ/action/storage_attestation","attest_author":"https://pith.science/pith/BQNNVBBJ34AFQVJ6U75OZELHZJ/action/author_attestation","sign_citation":"https://pith.science/pith/BQNNVBBJ34AFQVJ6U75OZELHZJ/action/citation_signature","submit_replication":"https://pith.science/pith/BQNNVBBJ34AFQVJ6U75OZELHZJ/action/replication_record"}},"created_at":"2026-05-22T01:03:20.038184+00:00","updated_at":"2026-05-22T01:03:20.038184+00:00"}