{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:5ZJGW6YGUSKYDAVITH4G7J2GZ3","short_pith_number":"pith:5ZJGW6YG","canonical_record":{"source":{"id":"2503.11557","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-14T16:26:11Z","cross_cats_sorted":[],"title_canon_sha256":"d491e7e57f89288c3301bfeba8a2574f96653cafef34f2a8d153b4186e9ddc77","abstract_canon_sha256":"9651a3dc3a151467d24f6f636bb00ade489d9b76b00080497fa1df50673aed40"},"schema_version":"1.0"},"canonical_sha256":"ee526b7b06a4958182a899f86fa746cecbd24dca364640c83a9633c4fde5c3d1","source":{"kind":"arxiv","id":"2503.11557","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.11557","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"arxiv_version","alias_value":"2503.11557v1","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.11557","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"pith_short_12","alias_value":"5ZJGW6YGUSKY","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"pith_short_16","alias_value":"5ZJGW6YGUSKYDAVI","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"pith_short_8","alias_value":"5ZJGW6YG","created_at":"2026-07-05T10:31:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:5ZJGW6YGUSKYDAVITH4G7J2GZ3","target":"record","payload":{"canonical_record":{"source":{"id":"2503.11557","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-14T16:26:11Z","cross_cats_sorted":[],"title_canon_sha256":"d491e7e57f89288c3301bfeba8a2574f96653cafef34f2a8d153b4186e9ddc77","abstract_canon_sha256":"9651a3dc3a151467d24f6f636bb00ade489d9b76b00080497fa1df50673aed40"},"schema_version":"1.0"},"canonical_sha256":"ee526b7b06a4958182a899f86fa746cecbd24dca364640c83a9633c4fde5c3d1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:31:34.547298Z","signature_b64":"al9zH9KqxuxdU1F30Z7NNvwormWqeSzNbyLZYg3p4njxs8hTE2oyW3hePUqvEE4+ZkiMQ4Mn8W3iqZ3JL9G0DQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ee526b7b06a4958182a899f86fa746cecbd24dca364640c83a9633c4fde5c3d1","last_reissued_at":"2026-07-05T10:31:34.546745Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:31:34.546745Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.11557","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:31:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8+dDqCyMWGOi1gpL+oxq5ePQ6kH8Xi/5kOUGWCbFf9HzdC4/ipKETFnTZ3H42bFklUZ6lIEiNxXHpaVfjpInCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:52:37.693617Z"},"content_sha256":"084a7b769aa3c030102452c355f3e308b57e06d80e3c01be962b20e6ec1a70aa","schema_version":"1.0","event_id":"sha256:084a7b769aa3c030102452c355f3e308b57e06d80e3c01be962b20e6ec1a70aa"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:5ZJGW6YGUSKYDAVITH4G7J2GZ3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"VERIFY: A Benchmark of Visual Explanation and Reasoning for Investigating Multimodal Reasoning Fidelity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ali Vosoughi, Chen Chen, Chenliang Xu, Guangyu Sun, Jiarui Wu, Jing Bi, Jinxi He, JunJia Guo, Luchuan Song, Susan Liang, Yunlong Tang","submitted_at":"2025-03-14T16:26:11Z","abstract_excerpt":"Visual reasoning is central to human cognition, enabling individuals to interpret and abstractly understand their environment. Although recent Multimodal Large Language Models (MLLMs) have demonstrated impressive performance across language and vision-language tasks, existing benchmarks primarily measure recognition-based skills and inadequately assess true visual reasoning capabilities. To bridge this critical gap, we introduce VERIFY, a benchmark explicitly designed to isolate and rigorously evaluate the visual reasoning capabilities of state-of-the-art MLLMs. VERIFY compels models to reason"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.11557","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/2503.11557/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T10:31:34Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S993JA9GBToI/Lo3wxXDvVbXldQ1nl+OolnEqax0jdKrhDIUnb+Kh5xcP9t+FgPVk+gMhJl6wD11G1U6D8p0BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:52:37.694673Z"},"content_sha256":"d5eeb2a4e124bcdfd912f6122e011d337e890a544404bd2408b27482e24e3d3f","schema_version":"1.0","event_id":"sha256:d5eeb2a4e124bcdfd912f6122e011d337e890a544404bd2408b27482e24e3d3f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/5ZJGW6YGUSKYDAVITH4G7J2GZ3/bundle.json","state_url":"https://pith.science/pith/5ZJGW6YGUSKYDAVITH4G7J2GZ3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/5ZJGW6YGUSKYDAVITH4G7J2GZ3/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-04T16:52:37Z","links":{"resolver":"https://pith.science/pith/5ZJGW6YGUSKYDAVITH4G7J2GZ3","bundle":"https://pith.science/pith/5ZJGW6YGUSKYDAVITH4G7J2GZ3/bundle.json","state":"https://pith.science/pith/5ZJGW6YGUSKYDAVITH4G7J2GZ3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/5ZJGW6YGUSKYDAVITH4G7J2GZ3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:5ZJGW6YGUSKYDAVITH4G7J2GZ3","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":"9651a3dc3a151467d24f6f636bb00ade489d9b76b00080497fa1df50673aed40","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-14T16:26:11Z","title_canon_sha256":"d491e7e57f89288c3301bfeba8a2574f96653cafef34f2a8d153b4186e9ddc77"},"schema_version":"1.0","source":{"id":"2503.11557","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.11557","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"arxiv_version","alias_value":"2503.11557v1","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.11557","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"pith_short_12","alias_value":"5ZJGW6YGUSKY","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"pith_short_16","alias_value":"5ZJGW6YGUSKYDAVI","created_at":"2026-07-05T10:31:34Z"},{"alias_kind":"pith_short_8","alias_value":"5ZJGW6YG","created_at":"2026-07-05T10:31:34Z"}],"graph_snapshots":[{"event_id":"sha256:d5eeb2a4e124bcdfd912f6122e011d337e890a544404bd2408b27482e24e3d3f","target":"graph","created_at":"2026-07-05T10:31:34Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2503.11557/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Visual reasoning is central to human cognition, enabling individuals to interpret and abstractly understand their environment. Although recent Multimodal Large Language Models (MLLMs) have demonstrated impressive performance across language and vision-language tasks, existing benchmarks primarily measure recognition-based skills and inadequately assess true visual reasoning capabilities. To bridge this critical gap, we introduce VERIFY, a benchmark explicitly designed to isolate and rigorously evaluate the visual reasoning capabilities of state-of-the-art MLLMs. VERIFY compels models to reason","authors_text":"Ali Vosoughi, Chen Chen, Chenliang Xu, Guangyu Sun, Jiarui Wu, Jing Bi, Jinxi He, JunJia Guo, Luchuan Song, Susan Liang, Yunlong Tang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-14T16:26:11Z","title":"VERIFY: A Benchmark of Visual Explanation and Reasoning for Investigating Multimodal Reasoning Fidelity"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.11557","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:084a7b769aa3c030102452c355f3e308b57e06d80e3c01be962b20e6ec1a70aa","target":"record","created_at":"2026-07-05T10:31:34Z","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":"9651a3dc3a151467d24f6f636bb00ade489d9b76b00080497fa1df50673aed40","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-14T16:26:11Z","title_canon_sha256":"d491e7e57f89288c3301bfeba8a2574f96653cafef34f2a8d153b4186e9ddc77"},"schema_version":"1.0","source":{"id":"2503.11557","kind":"arxiv","version":1}},"canonical_sha256":"ee526b7b06a4958182a899f86fa746cecbd24dca364640c83a9633c4fde5c3d1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ee526b7b06a4958182a899f86fa746cecbd24dca364640c83a9633c4fde5c3d1","first_computed_at":"2026-07-05T10:31:34.546745Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:31:34.546745Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"al9zH9KqxuxdU1F30Z7NNvwormWqeSzNbyLZYg3p4njxs8hTE2oyW3hePUqvEE4+ZkiMQ4Mn8W3iqZ3JL9G0DQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:31:34.547298Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.11557","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:084a7b769aa3c030102452c355f3e308b57e06d80e3c01be962b20e6ec1a70aa","sha256:d5eeb2a4e124bcdfd912f6122e011d337e890a544404bd2408b27482e24e3d3f"],"state_sha256":"fb9c35bb92113dc56b65a4608e1d654e4a00be755006efdfd74ca4e449b8e8d5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sdYjcDkyL/t7/dw8bQikl5XQxpUyaXmkptn+ysgfHBYm7G7r6JlfVgh6Z3ciK3AcXwcC2xTk277tUVQ+mEiyBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T16:52:37.701594Z","bundle_sha256":"65bd86ef08fd4eea4c5436d2f36044f298655e89951d79efd1f94607dc05508f"}}