{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:V57FGAYEXYXIVZNSSRSPQACGCK","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":"7ced050946093725dee4fef110dfe64123afee707c193ea81452b83ac304defd","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-21T07:09:58Z","title_canon_sha256":"b017e794784709853a18e8cfa24b9e803a6fb39d92a750a23dcc63111d3006ce"},"schema_version":"1.0","source":{"id":"2402.15527","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.15527","created_at":"2026-07-05T07:48:54Z"},{"alias_kind":"arxiv_version","alias_value":"2402.15527v1","created_at":"2026-07-05T07:48:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.15527","created_at":"2026-07-05T07:48:54Z"},{"alias_kind":"pith_short_12","alias_value":"V57FGAYEXYXI","created_at":"2026-07-05T07:48:54Z"},{"alias_kind":"pith_short_16","alias_value":"V57FGAYEXYXIVZNS","created_at":"2026-07-05T07:48:54Z"},{"alias_kind":"pith_short_8","alias_value":"V57FGAYE","created_at":"2026-07-05T07:48:54Z"}],"graph_snapshots":[{"event_id":"sha256:aab7b2f92850a0344e06a4244fd8d3fba09ac4466c4f14363b59dfee9fe94831","target":"graph","created_at":"2026-07-05T07:48:54Z","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/2402.15527/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present PCA-Bench, a multimodal decision-making benchmark for evaluating the integrated capabilities of Multimodal Large Language Models (MLLMs). Departing from previous benchmarks focusing on simplistic tasks and individual model capability, PCA-Bench introduces three complex scenarios: autonomous driving, domestic robotics, and open-world games. Given task instructions and diverse contexts, the model is required to seamlessly integrate multiple capabilities of Perception, Cognition, and Action in a reasoning chain to make accurate decisions. Moreover, PCA-Bench features error localization","authors_text":"Baobao Chang, Haozhe Zhao, Liang Chen, Peiyi Wang, Shuhuai Ren, Tianyu Liu, Xiangdi Meng, Yichi Zhang, Yuchi Wang, Zefan Cai","cross_cats":["cs.AI","cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-21T07:09:58Z","title":"PCA-Bench: Evaluating Multimodal Large Language Models in Perception-Cognition-Action Chain"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.15527","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:ca3fef71b03f0325ae4948d20760bdafd40eb6ee51b303dc29ff89ef737a4093","target":"record","created_at":"2026-07-05T07:48:54Z","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":"7ced050946093725dee4fef110dfe64123afee707c193ea81452b83ac304defd","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-02-21T07:09:58Z","title_canon_sha256":"b017e794784709853a18e8cfa24b9e803a6fb39d92a750a23dcc63111d3006ce"},"schema_version":"1.0","source":{"id":"2402.15527","kind":"arxiv","version":1}},"canonical_sha256":"af7e530304be2e8ae5b29464f800461280066bf4a939e0313ce3c220be9d01c7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"af7e530304be2e8ae5b29464f800461280066bf4a939e0313ce3c220be9d01c7","first_computed_at":"2026-07-05T07:48:54.276600Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:48:54.276600Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3Iq+Rx91NM+yzciz6KMLG53OBVxR7cBbXepKJIyHThs18GR69WBO3G18GRx+fMBc9P2VFCSWdDl4W0ah1B6xAg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:48:54.277117Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.15527","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ca3fef71b03f0325ae4948d20760bdafd40eb6ee51b303dc29ff89ef737a4093","sha256:aab7b2f92850a0344e06a4244fd8d3fba09ac4466c4f14363b59dfee9fe94831"],"state_sha256":"e366a544077b42ff2c80be60211c5223a8023075d3a3da5606cf56261dfada5b"}