{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:WPVZWI7A2G7D42XIMERHX4N6EL","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":"e31c0ca9f328c39774b68c2acf34450f220a26d966b69fd41762ea883a92f466","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-06-16T20:53:25Z","title_canon_sha256":"dcae5b41546b6ad524287916fd0080e7248e075d30d8b05a18e2f92fd2339c9e"},"schema_version":"1.0","source":{"id":"2406.11069","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.11069","created_at":"2026-07-05T08:32:48Z"},{"alias_kind":"arxiv_version","alias_value":"2406.11069v1","created_at":"2026-07-05T08:32:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.11069","created_at":"2026-07-05T08:32:48Z"},{"alias_kind":"pith_short_12","alias_value":"WPVZWI7A2G7D","created_at":"2026-07-05T08:32:48Z"},{"alias_kind":"pith_short_16","alias_value":"WPVZWI7A2G7D42XI","created_at":"2026-07-05T08:32:48Z"},{"alias_kind":"pith_short_8","alias_value":"WPVZWI7A","created_at":"2026-07-05T08:32:48Z"}],"graph_snapshots":[{"event_id":"sha256:c206a6cd4b4080e6d25ce8df4cb8be1bdd2738fde3ee91ba20e0ff02c1148ab4","target":"graph","created_at":"2026-07-05T08:32:48Z","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/2406.11069/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent breakthroughs in vision-language models (VLMs) emphasize the necessity of benchmarking human preferences in real-world multimodal interactions. To address this gap, we launched WildVision-Arena (WV-Arena), an online platform that collects human preferences to evaluate VLMs. We curated WV-Bench by selecting 500 high-quality samples from 8,000 user submissions in WV-Arena. WV-Bench uses GPT-4 as the judge to compare each VLM with Claude-3-Sonnet, achieving a Spearman correlation of 0.94 with the WV-Arena Elo. This significantly outperforms other benchmarks like MMVet, MMMU, and MMStar.\n  ","authors_text":"Bill Yuchen Lin, Dongfu Jiang, Wenhu Chen, William Yang Wang, Yejin Choi, Yujie Lu","cross_cats":["cs.AI","cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-06-16T20:53:25Z","title":"WildVision: Evaluating Vision-Language Models in the Wild with Human Preferences"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.11069","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":"sha256:45e7d326b05d71c70dc55e73b566cb88abf6da56ee9da2767bc337f0c8e273d9","target":"record","created_at":"2026-07-05T08:32:48Z","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":"e31c0ca9f328c39774b68c2acf34450f220a26d966b69fd41762ea883a92f466","cross_cats_sorted":["cs.AI","cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2024-06-16T20:53:25Z","title_canon_sha256":"dcae5b41546b6ad524287916fd0080e7248e075d30d8b05a18e2f92fd2339c9e"},"schema_version":"1.0","source":{"id":"2406.11069","kind":"arxiv","version":1}},"canonical_sha256":"b3eb9b23e0d1be3e6ae861227bf1be22fb37c7e706c318ffdc2d61ee96ffab67","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b3eb9b23e0d1be3e6ae861227bf1be22fb37c7e706c318ffdc2d61ee96ffab67","first_computed_at":"2026-07-05T08:32:48.513429Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:32:48.513429Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"jNL2vkZ0dAqMzprHVWvEgosvRoiKu15jYVef8sGNHNrIhqofeizkcZCgDPxiqw7n7Q8Zz7K2WueRIJdSqgDgCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:32:48.513840Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.11069","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:45e7d326b05d71c70dc55e73b566cb88abf6da56ee9da2767bc337f0c8e273d9","sha256:c206a6cd4b4080e6d25ce8df4cb8be1bdd2738fde3ee91ba20e0ff02c1148ab4"],"state_sha256":"364221256d0a8f34d3826a8faecfb697ecca02d971faa631dd8bcc400193c5c4"}