{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:3ALO7SHLEHDZVESOJ3L3VEO2JO","short_pith_number":"pith:3ALO7SHL","schema_version":"1.0","canonical_sha256":"d816efc8eb21c79a924e4ed7ba91da4bbe9fb9d001f245d66c595011614c3cca","source":{"kind":"arxiv","id":"2606.02357","version":1},"attestation_state":"computed","paper":{"title":"Do Multimodal Agents Really Benefit from Tool Use? A Systematic Study of Capability Gains","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Donglei Yu, Garvin Guo, Huaxing Liu, Minpeng Liao, Qinghao Wang, Shuai Li, Xiang Wang, Xinpei Zhao, Yu Chen","submitted_at":"2026-06-01T15:04:25Z","abstract_excerpt":"Tool-augmented multimodal agents show strong benchmark gains, often taken as evidence that agents have learned to use tools. We argue that this interpretation can be premature: a tool-call trace alone does not show whether the tool supplied answer-critical information. We study two representative ``thinking with images'' agents, Thyme and DeepEyesV2, across real-world understanding, OCR, chart understanding, and mathematical reasoning. Each agent is compared with its Tool-Free counterpart and with a Pure-Text Reasoner trained from the same source pool without tool-calling trajectories. Tool ac"},"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":"2606.02357","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-01T15:04:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"9a30e33267fa977dcff202ab77f829c2316a245dfa59ee2a766750353ed4190b","abstract_canon_sha256":"4247243578964ae049b2c7f1db8df0f9b4dccc884c799287e244b245ce690ba6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T03:04:57.040628Z","signature_b64":"lJnwsb+pYouibql1zbpUPRD7bp7GsPP5cpHxCq/7mHdYQRz1KH5mrK/+HvXNiCS14Ufn0hsIgvwPPs3fsu+3Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d816efc8eb21c79a924e4ed7ba91da4bbe9fb9d001f245d66c595011614c3cca","last_reissued_at":"2026-06-02T03:04:57.040238Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T03:04:57.040238Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do Multimodal Agents Really Benefit from Tool Use? A Systematic Study of Capability Gains","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Donglei Yu, Garvin Guo, Huaxing Liu, Minpeng Liao, Qinghao Wang, Shuai Li, Xiang Wang, Xinpei Zhao, Yu Chen","submitted_at":"2026-06-01T15:04:25Z","abstract_excerpt":"Tool-augmented multimodal agents show strong benchmark gains, often taken as evidence that agents have learned to use tools. We argue that this interpretation can be premature: a tool-call trace alone does not show whether the tool supplied answer-critical information. We study two representative ``thinking with images'' agents, Thyme and DeepEyesV2, across real-world understanding, OCR, chart understanding, and mathematical reasoning. Each agent is compared with its Tool-Free counterpart and with a Pure-Text Reasoner trained from the same source pool without tool-calling trajectories. Tool ac"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.02357","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/2606.02357/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":"2606.02357","created_at":"2026-06-02T03:04:57.040293+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.02357v1","created_at":"2026-06-02T03:04:57.040293+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.02357","created_at":"2026-06-02T03:04:57.040293+00:00"},{"alias_kind":"pith_short_12","alias_value":"3ALO7SHLEHDZ","created_at":"2026-06-02T03:04:57.040293+00:00"},{"alias_kind":"pith_short_16","alias_value":"3ALO7SHLEHDZVESO","created_at":"2026-06-02T03:04:57.040293+00:00"},{"alias_kind":"pith_short_8","alias_value":"3ALO7SHL","created_at":"2026-06-02T03:04:57.040293+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/3ALO7SHLEHDZVESOJ3L3VEO2JO","json":"https://pith.science/pith/3ALO7SHLEHDZVESOJ3L3VEO2JO.json","graph_json":"https://pith.science/api/pith-number/3ALO7SHLEHDZVESOJ3L3VEO2JO/graph.json","events_json":"https://pith.science/api/pith-number/3ALO7SHLEHDZVESOJ3L3VEO2JO/events.json","paper":"https://pith.science/paper/3ALO7SHL"},"agent_actions":{"view_html":"https://pith.science/pith/3ALO7SHLEHDZVESOJ3L3VEO2JO","download_json":"https://pith.science/pith/3ALO7SHLEHDZVESOJ3L3VEO2JO.json","view_paper":"https://pith.science/paper/3ALO7SHL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.02357&json=true","fetch_graph":"https://pith.science/api/pith-number/3ALO7SHLEHDZVESOJ3L3VEO2JO/graph.json","fetch_events":"https://pith.science/api/pith-number/3ALO7SHLEHDZVESOJ3L3VEO2JO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3ALO7SHLEHDZVESOJ3L3VEO2JO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3ALO7SHLEHDZVESOJ3L3VEO2JO/action/storage_attestation","attest_author":"https://pith.science/pith/3ALO7SHLEHDZVESOJ3L3VEO2JO/action/author_attestation","sign_citation":"https://pith.science/pith/3ALO7SHLEHDZVESOJ3L3VEO2JO/action/citation_signature","submit_replication":"https://pith.science/pith/3ALO7SHLEHDZVESOJ3L3VEO2JO/action/replication_record"}},"created_at":"2026-06-02T03:04:57.040293+00:00","updated_at":"2026-06-02T03:04:57.040293+00:00"}