{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JALZQVRK6ZK3HLEKQMFZJNUHEF","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":"5ac7bf2e46dfbc5996b4f1eedebd8ee9ff2e6e9a0719a7cba0155b8f39ebc555","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-02T07:27:58Z","title_canon_sha256":"92410496320bd041e903a8d24666b7aaf7480725af52c43c2f917bb7dff81253"},"schema_version":"1.0","source":{"id":"2607.01813","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.01813","created_at":"2026-07-03T01:17:30Z"},{"alias_kind":"arxiv_version","alias_value":"2607.01813v1","created_at":"2026-07-03T01:17:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.01813","created_at":"2026-07-03T01:17:30Z"},{"alias_kind":"pith_short_12","alias_value":"JALZQVRK6ZK3","created_at":"2026-07-03T01:17:30Z"},{"alias_kind":"pith_short_16","alias_value":"JALZQVRK6ZK3HLEK","created_at":"2026-07-03T01:17:30Z"},{"alias_kind":"pith_short_8","alias_value":"JALZQVRK","created_at":"2026-07-03T01:17:30Z"}],"graph_snapshots":[{"event_id":"sha256:872c345273b9b7013b3eca6d718d6ce7d46162765f5aa60b846a62d3dfb9f30a","target":"graph","created_at":"2026-07-03T01:17:30Z","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/2607.01813/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluation benchmarks are essential for assessing vision-language models (VLMs), but most multimodal benchmarks are static, making them vulnerable to temporal staleness, data contamination, and costly maintenance. We present MMBench-Live, a continuously evolving multimodal benchmark built by a multi-agent-driven automated pipeline. Our framework treats benchmark evolution as task-guided dataset construction, integrating structured benchmark specification, feedback-controlled real-time data acquisition, and verifiable QA generation with executable reasoning. To maintain cross-version comparabil","authors_text":"Bo Zhou, Kongming Liang, Shousheng Zhao, Yuanzhi Liu, Zhanyu Ma","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-02T07:27:58Z","title":"MMBench-Live: A Continuously Evolving Benchmark for Multimodal Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.01813","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":"sha256:872b8b008b93d5d7939052bc7ed64dd5358bb84651b3528715f5189f0d24145f","target":"record","created_at":"2026-07-03T01:17:30Z","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":"5ac7bf2e46dfbc5996b4f1eedebd8ee9ff2e6e9a0719a7cba0155b8f39ebc555","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-02T07:27:58Z","title_canon_sha256":"92410496320bd041e903a8d24666b7aaf7480725af52c43c2f917bb7dff81253"},"schema_version":"1.0","source":{"id":"2607.01813","kind":"arxiv","version":1}},"canonical_sha256":"481798562af655b3ac8a830b94b6872149155d8e50daeb7167ba702810700a57","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"481798562af655b3ac8a830b94b6872149155d8e50daeb7167ba702810700a57","first_computed_at":"2026-07-03T01:17:30.590743Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-03T01:17:30.590743Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Y6XhPXJ9zmAkLt0N54Wa3csYiln2o4jwiZKx0VPS1B7HEbnQqjUHE+s234fRIGu8H4QN//nXeYJhdUbibCIECg==","signature_status":"signed_v1","signed_at":"2026-07-03T01:17:30.591258Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.01813","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:872b8b008b93d5d7939052bc7ed64dd5358bb84651b3528715f5189f0d24145f","sha256:872c345273b9b7013b3eca6d718d6ce7d46162765f5aa60b846a62d3dfb9f30a"],"state_sha256":"218d139f61e26648187e3e199070608c4a9879b73e58f1dfed96fcc711490e29"}