{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:U4QCKXJFQT35IAFVW7WD5YPL4V","short_pith_number":"pith:U4QCKXJF","schema_version":"1.0","canonical_sha256":"a720255d2584f7d400b5b7ec3ee1ebe573ea398e0d119150b700a42acb7e3656","source":{"kind":"arxiv","id":"2507.23135","version":1},"attestation_state":"computed","paper":{"title":"ISO-Bench: Benchmarking Multimodal Causal Reasoning in Visual-Language Models through Procedural Plans","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ananya Sadana, Jiawei Zhou, Yash Kumar Lal","submitted_at":"2025-07-30T22:30:48Z","abstract_excerpt":"Understanding causal relationships across modalities is a core challenge for multimodal models operating in real-world environments. We introduce ISO-Bench, a benchmark for evaluating whether models can infer causal dependencies between visual observations and procedural text. Each example presents an image of a task step and a text snippet from a plan, with the goal of deciding whether the visual step occurs before or after the referenced text step. Evaluation results on ten frontier vision-language models show underwhelming performance: the best zero-shot F1 is only 0.57, and chain-of-though"},"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":"2507.23135","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-07-30T22:30:48Z","cross_cats_sorted":[],"title_canon_sha256":"3a974f9733f834e69ce501fa086dba22ab160324921bc1d9992ec51fbcf46fc8","abstract_canon_sha256":"1a58a733dfaa265bc43b7d3626992a5b278a7891f84d9bb5a2e5bfafb354d933"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:46:06.255605Z","signature_b64":"T4gcNVgQkl36lNW1Vq9V6C4+Qd/58bAWBaHHaM+qbg21p3cTQejzSCvVLFHEJV/xRXXDsOYlXVfhrkFpZOf4DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a720255d2584f7d400b5b7ec3ee1ebe573ea398e0d119150b700a42acb7e3656","last_reissued_at":"2026-07-05T11:46:06.255053Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:46:06.255053Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ISO-Bench: Benchmarking Multimodal Causal Reasoning in Visual-Language Models through Procedural Plans","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ananya Sadana, Jiawei Zhou, Yash Kumar Lal","submitted_at":"2025-07-30T22:30:48Z","abstract_excerpt":"Understanding causal relationships across modalities is a core challenge for multimodal models operating in real-world environments. We introduce ISO-Bench, a benchmark for evaluating whether models can infer causal dependencies between visual observations and procedural text. Each example presents an image of a task step and a text snippet from a plan, with the goal of deciding whether the visual step occurs before or after the referenced text step. Evaluation results on ten frontier vision-language models show underwhelming performance: the best zero-shot F1 is only 0.57, and chain-of-though"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2507.23135","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/2507.23135/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":"2507.23135","created_at":"2026-07-05T11:46:06.255121+00:00"},{"alias_kind":"arxiv_version","alias_value":"2507.23135v1","created_at":"2026-07-05T11:46:06.255121+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2507.23135","created_at":"2026-07-05T11:46:06.255121+00:00"},{"alias_kind":"pith_short_12","alias_value":"U4QCKXJFQT35","created_at":"2026-07-05T11:46:06.255121+00:00"},{"alias_kind":"pith_short_16","alias_value":"U4QCKXJFQT35IAFV","created_at":"2026-07-05T11:46:06.255121+00:00"},{"alias_kind":"pith_short_8","alias_value":"U4QCKXJF","created_at":"2026-07-05T11:46:06.255121+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04479","citing_title":"Evaluating Reasoning Fidelity in Visual Text Generation","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22903","citing_title":"Seeing without Looking: Do Vision-Language Benchmarks Really Test Vision?","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/U4QCKXJFQT35IAFVW7WD5YPL4V","json":"https://pith.science/pith/U4QCKXJFQT35IAFVW7WD5YPL4V.json","graph_json":"https://pith.science/api/pith-number/U4QCKXJFQT35IAFVW7WD5YPL4V/graph.json","events_json":"https://pith.science/api/pith-number/U4QCKXJFQT35IAFVW7WD5YPL4V/events.json","paper":"https://pith.science/paper/U4QCKXJF"},"agent_actions":{"view_html":"https://pith.science/pith/U4QCKXJFQT35IAFVW7WD5YPL4V","download_json":"https://pith.science/pith/U4QCKXJFQT35IAFVW7WD5YPL4V.json","view_paper":"https://pith.science/paper/U4QCKXJF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2507.23135&json=true","fetch_graph":"https://pith.science/api/pith-number/U4QCKXJFQT35IAFVW7WD5YPL4V/graph.json","fetch_events":"https://pith.science/api/pith-number/U4QCKXJFQT35IAFVW7WD5YPL4V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/U4QCKXJFQT35IAFVW7WD5YPL4V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/U4QCKXJFQT35IAFVW7WD5YPL4V/action/storage_attestation","attest_author":"https://pith.science/pith/U4QCKXJFQT35IAFVW7WD5YPL4V/action/author_attestation","sign_citation":"https://pith.science/pith/U4QCKXJFQT35IAFVW7WD5YPL4V/action/citation_signature","submit_replication":"https://pith.science/pith/U4QCKXJFQT35IAFVW7WD5YPL4V/action/replication_record"}},"created_at":"2026-07-05T11:46:06.255121+00:00","updated_at":"2026-07-05T11:46:06.255121+00:00"}