{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7IYHSNNLC4VW62KM3OMN25VGZA","short_pith_number":"pith:7IYHSNNL","schema_version":"1.0","canonical_sha256":"fa307935ab172b6f694cdb98dd76a6c8019ca0799668679f5d20485a569718f1","source":{"kind":"arxiv","id":"2503.17489","version":1},"attestation_state":"computed","paper":{"title":"Judge Anything: MLLM as a Judge Across Any Modality","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Dongping Chen, Guohao Wang, Philip S. Yu, Qi Qin, Shuang Gong, Shu Pu, Yaochen Wang, Yao Wan, Yi Gui, Yuhang Chen, Zetong Zhou, Zhiyuan Zhang, Zhongyi Zhang","submitted_at":"2025-03-21T18:59:20Z","abstract_excerpt":"Evaluating generative foundation models on open-ended multimodal understanding (MMU) and generation (MMG) tasks across diverse modalities (e.g., images, audio, video) poses significant challenges due to the complexity of cross-modal interactions. To this end, the idea of utilizing Multimodal LLMs (MLLMs) as automated judges has emerged, with encouraging results in assessing vision-language understanding tasks. Moving further, this paper extends MLLM-as-a-Judge across modalities to a unified manner by introducing two benchmarks, TaskAnything and JudgeAnything, to respectively evaluate the overa"},"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":"2503.17489","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-21T18:59:20Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"65d2e1c2b827315697b745a02e58c8c7e3e1e94836264fc43f1e0772622840fe","abstract_canon_sha256":"923bdd74d84f6da4129c3374e8e5ff4aacaaafd60807e4bf8fead47db8040c6d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:37:10.540769Z","signature_b64":"j9VUuEFgbY80B/g14UGqKreJkKryH7fVLJq+ZKoh0SBS6mjsAfOtQsvcQ/CkhS/g9ZJxDz5u92byorQ7ie8lCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fa307935ab172b6f694cdb98dd76a6c8019ca0799668679f5d20485a569718f1","last_reissued_at":"2026-07-05T10:37:10.540218Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:37:10.540218Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Judge Anything: MLLM as a Judge Across Any Modality","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Dongping Chen, Guohao Wang, Philip S. Yu, Qi Qin, Shuang Gong, Shu Pu, Yaochen Wang, Yao Wan, Yi Gui, Yuhang Chen, Zetong Zhou, Zhiyuan Zhang, Zhongyi Zhang","submitted_at":"2025-03-21T18:59:20Z","abstract_excerpt":"Evaluating generative foundation models on open-ended multimodal understanding (MMU) and generation (MMG) tasks across diverse modalities (e.g., images, audio, video) poses significant challenges due to the complexity of cross-modal interactions. To this end, the idea of utilizing Multimodal LLMs (MLLMs) as automated judges has emerged, with encouraging results in assessing vision-language understanding tasks. Moving further, this paper extends MLLM-as-a-Judge across modalities to a unified manner by introducing two benchmarks, TaskAnything and JudgeAnything, to respectively evaluate the overa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.17489","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/2503.17489/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":"2503.17489","created_at":"2026-07-05T10:37:10.540286+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.17489v1","created_at":"2026-07-05T10:37:10.540286+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.17489","created_at":"2026-07-05T10:37:10.540286+00:00"},{"alias_kind":"pith_short_12","alias_value":"7IYHSNNLC4VW","created_at":"2026-07-05T10:37:10.540286+00:00"},{"alias_kind":"pith_short_16","alias_value":"7IYHSNNLC4VW62KM","created_at":"2026-07-05T10:37:10.540286+00:00"},{"alias_kind":"pith_short_8","alias_value":"7IYHSNNL","created_at":"2026-07-05T10:37:10.540286+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23543","citing_title":"VeriEvol: Scaling Multimodal Mathematical Reasoning via Verifiable Evol-Instruct","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2601.05563","citing_title":"What's Left Unsaid? Detecting and Correcting Misleading Omissions in Multimodal News Previews","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20544","citing_title":"Evian: Towards Explainable Visual Instruction-tuning Data Auditing","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/7IYHSNNLC4VW62KM3OMN25VGZA","json":"https://pith.science/pith/7IYHSNNLC4VW62KM3OMN25VGZA.json","graph_json":"https://pith.science/api/pith-number/7IYHSNNLC4VW62KM3OMN25VGZA/graph.json","events_json":"https://pith.science/api/pith-number/7IYHSNNLC4VW62KM3OMN25VGZA/events.json","paper":"https://pith.science/paper/7IYHSNNL"},"agent_actions":{"view_html":"https://pith.science/pith/7IYHSNNLC4VW62KM3OMN25VGZA","download_json":"https://pith.science/pith/7IYHSNNLC4VW62KM3OMN25VGZA.json","view_paper":"https://pith.science/paper/7IYHSNNL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.17489&json=true","fetch_graph":"https://pith.science/api/pith-number/7IYHSNNLC4VW62KM3OMN25VGZA/graph.json","fetch_events":"https://pith.science/api/pith-number/7IYHSNNLC4VW62KM3OMN25VGZA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7IYHSNNLC4VW62KM3OMN25VGZA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7IYHSNNLC4VW62KM3OMN25VGZA/action/storage_attestation","attest_author":"https://pith.science/pith/7IYHSNNLC4VW62KM3OMN25VGZA/action/author_attestation","sign_citation":"https://pith.science/pith/7IYHSNNLC4VW62KM3OMN25VGZA/action/citation_signature","submit_replication":"https://pith.science/pith/7IYHSNNLC4VW62KM3OMN25VGZA/action/replication_record"}},"created_at":"2026-07-05T10:37:10.540286+00:00","updated_at":"2026-07-05T10:37:10.540286+00:00"}