{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DTXPYG3FZ7DJH2HEIXTAKUV5E5","short_pith_number":"pith:DTXPYG3F","schema_version":"1.0","canonical_sha256":"1ceefc1b65cfc693e8e445e60552bd2773e09959db25ece738e6c6648994327b","source":{"kind":"arxiv","id":"2508.15802","version":1},"attestation_state":"computed","paper":{"title":"MAC: A Live Benchmark for Multimodal Large Language Models in Scientific Understanding","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dequan Wang, Jiahao Zhan, Jin Gao, Mohan Jiang","submitted_at":"2025-08-14T01:22:57Z","abstract_excerpt":"As multimodal large language models (MLLMs) grow increasingly capable, fixed benchmarks are gradually losing their effectiveness in evaluating high-level scientific understanding. In this paper, we introduce the Multimodal Academic Cover benchmark (MAC), a live benchmark that could continuously evolve with scientific advancement and model progress. MAC leverages over 25,000 image-text pairs sourced from issues of top-tier scientific journals such as Nature, Science, and Cell, challenging MLLMs to reason across abstract visual and textual scientific content. Experiments on our most recent yearl"},"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":"2508.15802","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-08-14T01:22:57Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ef977ae26d33a43543b6d9bfc276003a713777a03f4df16f4f6955b643b7d0b8","abstract_canon_sha256":"32d8dd92d86fda4fcacdac3769c1a12a6b3d449bddad2da04c9872a4542025da"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:57:21.307594Z","signature_b64":"X2BQiezXjueZbA/XOr/ngjlzHuWNfYGVZ6dWZhf21HrvP0HeyncwlzVDhYeZEgmsqUg7d9XUvVlxQLKnX+p2Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1ceefc1b65cfc693e8e445e60552bd2773e09959db25ece738e6c6648994327b","last_reissued_at":"2026-07-05T11:57:21.307177Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:57:21.307177Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAC: A Live Benchmark for Multimodal Large Language Models in Scientific Understanding","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dequan Wang, Jiahao Zhan, Jin Gao, Mohan Jiang","submitted_at":"2025-08-14T01:22:57Z","abstract_excerpt":"As multimodal large language models (MLLMs) grow increasingly capable, fixed benchmarks are gradually losing their effectiveness in evaluating high-level scientific understanding. In this paper, we introduce the Multimodal Academic Cover benchmark (MAC), a live benchmark that could continuously evolve with scientific advancement and model progress. MAC leverages over 25,000 image-text pairs sourced from issues of top-tier scientific journals such as Nature, Science, and Cell, challenging MLLMs to reason across abstract visual and textual scientific content. Experiments on our most recent yearl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.15802","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/2508.15802/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":"2508.15802","created_at":"2026-07-05T11:57:21.307230+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.15802v1","created_at":"2026-07-05T11:57:21.307230+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.15802","created_at":"2026-07-05T11:57:21.307230+00:00"},{"alias_kind":"pith_short_12","alias_value":"DTXPYG3FZ7DJ","created_at":"2026-07-05T11:57:21.307230+00:00"},{"alias_kind":"pith_short_16","alias_value":"DTXPYG3FZ7DJH2HE","created_at":"2026-07-05T11:57:21.307230+00:00"},{"alias_kind":"pith_short_8","alias_value":"DTXPYG3F","created_at":"2026-07-05T11:57:21.307230+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26196","citing_title":"From Structure to Synergy: A Survey of Vision-Language Perception Paradigm Evolution in Multimodal Large Language Models","ref_index":223,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DTXPYG3FZ7DJH2HEIXTAKUV5E5","json":"https://pith.science/pith/DTXPYG3FZ7DJH2HEIXTAKUV5E5.json","graph_json":"https://pith.science/api/pith-number/DTXPYG3FZ7DJH2HEIXTAKUV5E5/graph.json","events_json":"https://pith.science/api/pith-number/DTXPYG3FZ7DJH2HEIXTAKUV5E5/events.json","paper":"https://pith.science/paper/DTXPYG3F"},"agent_actions":{"view_html":"https://pith.science/pith/DTXPYG3FZ7DJH2HEIXTAKUV5E5","download_json":"https://pith.science/pith/DTXPYG3FZ7DJH2HEIXTAKUV5E5.json","view_paper":"https://pith.science/paper/DTXPYG3F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.15802&json=true","fetch_graph":"https://pith.science/api/pith-number/DTXPYG3FZ7DJH2HEIXTAKUV5E5/graph.json","fetch_events":"https://pith.science/api/pith-number/DTXPYG3FZ7DJH2HEIXTAKUV5E5/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DTXPYG3FZ7DJH2HEIXTAKUV5E5/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DTXPYG3FZ7DJH2HEIXTAKUV5E5/action/storage_attestation","attest_author":"https://pith.science/pith/DTXPYG3FZ7DJH2HEIXTAKUV5E5/action/author_attestation","sign_citation":"https://pith.science/pith/DTXPYG3FZ7DJH2HEIXTAKUV5E5/action/citation_signature","submit_replication":"https://pith.science/pith/DTXPYG3FZ7DJH2HEIXTAKUV5E5/action/replication_record"}},"created_at":"2026-07-05T11:57:21.307230+00:00","updated_at":"2026-07-05T11:57:21.307230+00:00"}