{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZPEV5IVE2TWMEBPATDY45VBLY7","short_pith_number":"pith:ZPEV5IVE","schema_version":"1.0","canonical_sha256":"cbc95ea2a4d4ecc205e098f1ced42bc7cffe95f1c971ea4ce86bc80aa0771293","source":{"kind":"arxiv","id":"2606.03054","version":1},"attestation_state":"computed","paper":{"title":"ToolGate: Token-Efficient Pre-Call Control for Tool-Augmented Vision-Language Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Anjie Liu, Jun Wang, Yan Song, Zhixun Chen, Zhongwei Yu, Ziqin Gong","submitted_at":"2026-06-02T02:44:27Z","abstract_excerpt":"Tool-augmented vision-language agents can acquire external perceptual evidence through OCR, detection, segmentation, and other tools, but executing every proposed tool call is costly and sometimes unnecessary. We study the pre-call control problem: after a ReAct-style VLM agent proposes a perceptual tool call, should the call be executed, or skipped before its output enters the context? Across five benchmarks, we find that the baseline agent exhibits poor local selectivity: helpful and harmful calls occur at similar rates (11.8% vs. 9.9%), while most calls do not change the immediate forced-an"},"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":"2606.03054","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-02T02:44:27Z","cross_cats_sorted":[],"title_canon_sha256":"4b754e289f01abcfefc2769b3175442b43f4c6c9e01199e14065d0c59f6f7f74","abstract_canon_sha256":"34d4e667fbcea62268fa0adddf2cba6c146c1701f3b6e9b02abb81c518b13bd9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T01:05:30.438845Z","signature_b64":"7dIvuk7XPE+6KDSzJ0yWIITsnq0nKVjN0WnFhKV7YqGT7K0UcwmnDoMfKGsVX4tBk9xCCdRgLWJtrmeYBfIZDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cbc95ea2a4d4ecc205e098f1ced42bc7cffe95f1c971ea4ce86bc80aa0771293","last_reissued_at":"2026-06-03T01:05:30.438388Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T01:05:30.438388Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ToolGate: Token-Efficient Pre-Call Control for Tool-Augmented Vision-Language Agents","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Anjie Liu, Jun Wang, Yan Song, Zhixun Chen, Zhongwei Yu, Ziqin Gong","submitted_at":"2026-06-02T02:44:27Z","abstract_excerpt":"Tool-augmented vision-language agents can acquire external perceptual evidence through OCR, detection, segmentation, and other tools, but executing every proposed tool call is costly and sometimes unnecessary. We study the pre-call control problem: after a ReAct-style VLM agent proposes a perceptual tool call, should the call be executed, or skipped before its output enters the context? Across five benchmarks, we find that the baseline agent exhibits poor local selectivity: helpful and harmful calls occur at similar rates (11.8% vs. 9.9%), while most calls do not change the immediate forced-an"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03054","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/2606.03054/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":"2606.03054","created_at":"2026-06-03T01:05:30.438455+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03054v1","created_at":"2026-06-03T01:05:30.438455+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03054","created_at":"2026-06-03T01:05:30.438455+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZPEV5IVE2TWM","created_at":"2026-06-03T01:05:30.438455+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZPEV5IVE2TWMEBPA","created_at":"2026-06-03T01:05:30.438455+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZPEV5IVE","created_at":"2026-06-03T01:05:30.438455+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/ZPEV5IVE2TWMEBPATDY45VBLY7","json":"https://pith.science/pith/ZPEV5IVE2TWMEBPATDY45VBLY7.json","graph_json":"https://pith.science/api/pith-number/ZPEV5IVE2TWMEBPATDY45VBLY7/graph.json","events_json":"https://pith.science/api/pith-number/ZPEV5IVE2TWMEBPATDY45VBLY7/events.json","paper":"https://pith.science/paper/ZPEV5IVE"},"agent_actions":{"view_html":"https://pith.science/pith/ZPEV5IVE2TWMEBPATDY45VBLY7","download_json":"https://pith.science/pith/ZPEV5IVE2TWMEBPATDY45VBLY7.json","view_paper":"https://pith.science/paper/ZPEV5IVE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03054&json=true","fetch_graph":"https://pith.science/api/pith-number/ZPEV5IVE2TWMEBPATDY45VBLY7/graph.json","fetch_events":"https://pith.science/api/pith-number/ZPEV5IVE2TWMEBPATDY45VBLY7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZPEV5IVE2TWMEBPATDY45VBLY7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZPEV5IVE2TWMEBPATDY45VBLY7/action/storage_attestation","attest_author":"https://pith.science/pith/ZPEV5IVE2TWMEBPATDY45VBLY7/action/author_attestation","sign_citation":"https://pith.science/pith/ZPEV5IVE2TWMEBPATDY45VBLY7/action/citation_signature","submit_replication":"https://pith.science/pith/ZPEV5IVE2TWMEBPATDY45VBLY7/action/replication_record"}},"created_at":"2026-06-03T01:05:30.438455+00:00","updated_at":"2026-06-03T01:05:30.438455+00:00"}