{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:BQNNVBBJ34AFQVJ6U75OZELHZJ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"99cf9536e1412fc5ddbb1e5237a7790d01959c8a3d3582c5cbc580429ec1a4a7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-10T01:37:40Z","title_canon_sha256":"c8e106fb54f734422ae3421a324836d082fe177bb5f4288476ca9d9e935ce317"},"schema_version":"1.0","source":{"id":"2605.09252","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.09252","created_at":"2026-05-22T01:03:20Z"},{"alias_kind":"arxiv_version","alias_value":"2605.09252v2","created_at":"2026-05-22T01:03:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.09252","created_at":"2026-05-22T01:03:20Z"},{"alias_kind":"pith_short_12","alias_value":"BQNNVBBJ34AF","created_at":"2026-05-22T01:03:20Z"},{"alias_kind":"pith_short_16","alias_value":"BQNNVBBJ34AFQVJ6","created_at":"2026-05-22T01:03:20Z"},{"alias_kind":"pith_short_8","alias_value":"BQNNVBBJ","created_at":"2026-05-22T01:03:20Z"}],"graph_snapshots":[{"event_id":"sha256:d611cf2a13c7ba817bf278d1fd874718c77a0e4e7839be43d940c35336c5893f","target":"graph","created_at":"2026-05-22T01:03:20Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"LLM agents already encode whether a tool is needed in their hidden states before generating any output."}],"snapshot_sha256":"790595cfe58e92e1c16bab1597804a24d4faf98df0ff9cc03a92820cf7ab7ec1"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"77e33f81b3256ad82f36fe162221af29dfe2357269e3eaa2b9a73da7879b7424"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"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"}],"endpoint":"/pith/2605.09252/integrity.json","findings":[],"snapshot_sha256":"bb6d31f5a88a142fcb03c12860ccb3f5bda74f798277f0b4cea0675a0655eb38","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Chung-En Sun, Ge Yan, Linbo Liu, Tsui-Wei Weng, Zimo Wang","cross_cats":[],"headline":"LLM agents already encode whether a tool is needed in their hidden states before generating any output.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-10T01:37:40Z","title":"LLM Agents Already Know When to Call Tools -- Even Without Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.09252","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-12T02:37:05.851364Z","id":"b3a95362-a864-48cf-aafa-22af8d956278","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"LLM agents already encode whether a tool is needed in their hidden states before generating any output.","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.","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."}},"verdict_id":"b3a95362-a864-48cf-aafa-22af8d956278"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:3db33f20ca966f68c890645523a01c12c4f6b0a260920e54495e7fbc977b1a50","target":"record","created_at":"2026-05-22T01:03:20Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"99cf9536e1412fc5ddbb1e5237a7790d01959c8a3d3582c5cbc580429ec1a4a7","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-05-10T01:37:40Z","title_canon_sha256":"c8e106fb54f734422ae3421a324836d082fe177bb5f4288476ca9d9e935ce317"},"schema_version":"1.0","source":{"id":"2605.09252","kind":"arxiv","version":2}},"canonical_sha256":"0c1ada8429df0058553ea7faec9167ca61b65e1916dfdda3830ff08938747085","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0c1ada8429df0058553ea7faec9167ca61b65e1916dfdda3830ff08938747085","first_computed_at":"2026-05-22T01:03:20.038088Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:03:20.038088Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9sG8xjNt31JkWk6BQfHJ8jSykIE1bbmOpBSUsSZL2msYKbPqCs+EwsIzoDxDW4hFvR4vRFiflFhC9en6kfcnDg==","signature_status":"signed_v1","signed_at":"2026-05-22T01:03:20.038764Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.09252","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3db33f20ca966f68c890645523a01c12c4f6b0a260920e54495e7fbc977b1a50","sha256:d611cf2a13c7ba817bf278d1fd874718c77a0e4e7839be43d940c35336c5893f"],"state_sha256":"17991523913e5f16e0d41b2f28790f5a76aa4e4c2e784efbbab8284030a3efc5"}