{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:HIN4TTLBHX6MZKPDJPPNRCWIPF","short_pith_number":"pith:HIN4TTLB","schema_version":"1.0","canonical_sha256":"3a1bc9cd613dfccca9e34bded88ac8794a843d0477fcdcc592f5d0c3dea870d3","source":{"kind":"arxiv","id":"2606.19965","version":1},"attestation_state":"computed","paper":{"title":"ROSE: Benchmarking the Perception-to-Action Gap in Multimodal Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jie Ren, Keze Wang, Yihao Wang, Zijian He","submitted_at":"2026-06-18T09:05:48Z","abstract_excerpt":"Multimodal large language models (MLLMs) are increasingly expected to act on visual information, yet the same scene may require different actions under different task contexts. How reliably can a model turn the same visual evidence into the action required by the current context? To answer this question, we introduce \\textsc{ROSE} (\\textbf{R}eference-conditioned \\textbf{O}ddity and \\textbf{S}ymbolic \\textbf{E}xecution), a controlled benchmark that holds the visual scene fixed while varying region constraints and required symbolic outputs. Through coupled counting and coordinate-action tasks, \\"},"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":"2606.19965","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-18T09:05:48Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a3441c6445fc48166882afc7403cd9901168c99a65981c1c065b384a88a5ddd5","abstract_canon_sha256":"49eb4ecf5a30f4e9aa7d6c34209c6b37c81561a17a474bb92a0a6fd5e24aff0d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:12:59.353770Z","signature_b64":"Y6vQLjDcxQqYJp7EMtVBmudlrUd0LuAvsabHElDFQ1HkEB0vzD+RnYX6418EPG9D+aiBza2Uf1q2HKDdlCAPCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a1bc9cd613dfccca9e34bded88ac8794a843d0477fcdcc592f5d0c3dea870d3","last_reissued_at":"2026-06-19T16:12:59.353400Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:12:59.353400Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ROSE: Benchmarking the Perception-to-Action Gap in Multimodal Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Jie Ren, Keze Wang, Yihao Wang, Zijian He","submitted_at":"2026-06-18T09:05:48Z","abstract_excerpt":"Multimodal large language models (MLLMs) are increasingly expected to act on visual information, yet the same scene may require different actions under different task contexts. How reliably can a model turn the same visual evidence into the action required by the current context? To answer this question, we introduce \\textsc{ROSE} (\\textbf{R}eference-conditioned \\textbf{O}ddity and \\textbf{S}ymbolic \\textbf{E}xecution), a controlled benchmark that holds the visual scene fixed while varying region constraints and required symbolic outputs. Through coupled counting and coordinate-action tasks, \\"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.19965","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/2606.19965/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":"2606.19965","created_at":"2026-06-19T16:12:59.353468+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.19965v1","created_at":"2026-06-19T16:12:59.353468+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.19965","created_at":"2026-06-19T16:12:59.353468+00:00"},{"alias_kind":"pith_short_12","alias_value":"HIN4TTLBHX6M","created_at":"2026-06-19T16:12:59.353468+00:00"},{"alias_kind":"pith_short_16","alias_value":"HIN4TTLBHX6MZKPD","created_at":"2026-06-19T16:12:59.353468+00:00"},{"alias_kind":"pith_short_8","alias_value":"HIN4TTLB","created_at":"2026-06-19T16:12:59.353468+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/HIN4TTLBHX6MZKPDJPPNRCWIPF","json":"https://pith.science/pith/HIN4TTLBHX6MZKPDJPPNRCWIPF.json","graph_json":"https://pith.science/api/pith-number/HIN4TTLBHX6MZKPDJPPNRCWIPF/graph.json","events_json":"https://pith.science/api/pith-number/HIN4TTLBHX6MZKPDJPPNRCWIPF/events.json","paper":"https://pith.science/paper/HIN4TTLB"},"agent_actions":{"view_html":"https://pith.science/pith/HIN4TTLBHX6MZKPDJPPNRCWIPF","download_json":"https://pith.science/pith/HIN4TTLBHX6MZKPDJPPNRCWIPF.json","view_paper":"https://pith.science/paper/HIN4TTLB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.19965&json=true","fetch_graph":"https://pith.science/api/pith-number/HIN4TTLBHX6MZKPDJPPNRCWIPF/graph.json","fetch_events":"https://pith.science/api/pith-number/HIN4TTLBHX6MZKPDJPPNRCWIPF/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HIN4TTLBHX6MZKPDJPPNRCWIPF/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HIN4TTLBHX6MZKPDJPPNRCWIPF/action/storage_attestation","attest_author":"https://pith.science/pith/HIN4TTLBHX6MZKPDJPPNRCWIPF/action/author_attestation","sign_citation":"https://pith.science/pith/HIN4TTLBHX6MZKPDJPPNRCWIPF/action/citation_signature","submit_replication":"https://pith.science/pith/HIN4TTLBHX6MZKPDJPPNRCWIPF/action/replication_record"}},"created_at":"2026-06-19T16:12:59.353468+00:00","updated_at":"2026-06-19T16:12:59.353468+00:00"}