{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:NY2IQIFT4VOHNHGXZKFWWQEMWX","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":"50bb0a3cdb20350596c243a39bce26c31c7faf7f29a689b2d1736080272ddcfb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-06T17:52:01Z","title_canon_sha256":"a483d122fc6d21bf25aacb6b6ef7a5d276295f33cc7cf900a3f7e472b8af839d"},"schema_version":"1.0","source":{"id":"2411.04075","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.04075","created_at":"2026-07-05T09:32:00Z"},{"alias_kind":"arxiv_version","alias_value":"2411.04075v1","created_at":"2026-07-05T09:32:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.04075","created_at":"2026-07-05T09:32:00Z"},{"alias_kind":"pith_short_12","alias_value":"NY2IQIFT4VOH","created_at":"2026-07-05T09:32:00Z"},{"alias_kind":"pith_short_16","alias_value":"NY2IQIFT4VOHNHGX","created_at":"2026-07-05T09:32:00Z"},{"alias_kind":"pith_short_8","alias_value":"NY2IQIFT","created_at":"2026-07-05T09:32:00Z"}],"graph_snapshots":[{"event_id":"sha256:11d140a72e67dbe966e55221ea548c08c3f5810c5e7363bdf80ec843e6cb132b","target":"graph","created_at":"2026-07-05T09:32:00Z","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/2411.04075/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Existing benchmarks for evaluating foundation models mainly focus on single-document, text-only tasks. However, they often fail to fully capture the complexity of research workflows, which typically involve interpreting non-textual data and gathering information across multiple documents. To address this gap, we introduce M3SciQA, a multi-modal, multi-document scientific question answering benchmark designed for a more comprehensive evaluation of foundation models. M3SciQA consists of 1,452 expert-annotated questions spanning 70 natural language processing paper clusters, where each cluster re","authors_text":"Arman Cohan, Chuhan Li, Deyuan Li, Yilun Zhao, Yixin Liu, Ziyao Shangguan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-06T17:52:01Z","title":"M3SciQA: A Multi-Modal Multi-Document Scientific QA Benchmark for Evaluating Foundation Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.04075","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":"sha256:1daf1db962d4643cd954659d4468c977bb14476f263ae40dff861efa4db8b408","target":"record","created_at":"2026-07-05T09:32:00Z","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":"50bb0a3cdb20350596c243a39bce26c31c7faf7f29a689b2d1736080272ddcfb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-06T17:52:01Z","title_canon_sha256":"a483d122fc6d21bf25aacb6b6ef7a5d276295f33cc7cf900a3f7e472b8af839d"},"schema_version":"1.0","source":{"id":"2411.04075","kind":"arxiv","version":1}},"canonical_sha256":"6e348820b3e55c769cd7ca8b6b408cb5d920f40b24ac24aa99598cb8fbb4fc06","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6e348820b3e55c769cd7ca8b6b408cb5d920f40b24ac24aa99598cb8fbb4fc06","first_computed_at":"2026-07-05T09:32:00.444415Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:32:00.444415Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"O8xefH507jWgtXk6Hvi6NRg5lX50cmo4z6LxjqElROXirsLnHn7cTcBjv4s9GctyRwyICaxCtaEUXxVB0lIFBw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:32:00.444903Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.04075","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1daf1db962d4643cd954659d4468c977bb14476f263ae40dff861efa4db8b408","sha256:11d140a72e67dbe966e55221ea548c08c3f5810c5e7363bdf80ec843e6cb132b"],"state_sha256":"a03dc98d5182e5c3dcb3bb9b03d78057726620311d02e14b41883bdc01ba85a4"}