{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:LALRMIHLSIBTET4JWYJEJ3IC7U","short_pith_number":"pith:LALRMIHL","canonical_record":{"source":{"id":"2404.05091","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-07T22:16:50Z","cross_cats_sorted":[],"title_canon_sha256":"942e3acb4b66f699034e5bc3b550699a3256d63ae89f2f1d550cbcc984d38d81","abstract_canon_sha256":"5505e39cdf99c7d2a040a8573422e1df08ae36d96a7b78ff7013284008bf50af"},"schema_version":"1.0"},"canonical_sha256":"58171620eb9203324f89b61244ed02fd2a9ded58b04d7f3e2bf8d985577311f7","source":{"kind":"arxiv","id":"2404.05091","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.05091","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"arxiv_version","alias_value":"2404.05091v4","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.05091","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"pith_short_12","alias_value":"LALRMIHLSIBT","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"pith_short_16","alias_value":"LALRMIHLSIBTET4J","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"pith_short_8","alias_value":"LALRMIHL","created_at":"2026-07-05T08:38:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:LALRMIHLSIBTET4JWYJEJ3IC7U","target":"record","payload":{"canonical_record":{"source":{"id":"2404.05091","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-07T22:16:50Z","cross_cats_sorted":[],"title_canon_sha256":"942e3acb4b66f699034e5bc3b550699a3256d63ae89f2f1d550cbcc984d38d81","abstract_canon_sha256":"5505e39cdf99c7d2a040a8573422e1df08ae36d96a7b78ff7013284008bf50af"},"schema_version":"1.0"},"canonical_sha256":"58171620eb9203324f89b61244ed02fd2a9ded58b04d7f3e2bf8d985577311f7","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:38:58.958870Z","signature_b64":"hvSO62Kd0UfO4fU3/xDYiKuDgMYs/DrptxazYsQdqprbHF12t67VXqKBYGtk7jBP7tAgqakkosVQbo8oCoxICg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58171620eb9203324f89b61244ed02fd2a9ded58b04d7f3e2bf8d985577311f7","last_reissued_at":"2026-07-05T08:38:58.958390Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:38:58.958390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.05091","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:38:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DyQZjTppxAwyiKYIENE6pY/pkArzWe1mNjH8Fvy81cY4bOTRHX0gwtYoSugcqctq0RpIZc+hsuXLLycRi7/DDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T04:35:48.527343Z"},"content_sha256":"f63f4c04ebaabc35de376cf12194aac757bda0279d4cc7b9bcb94ff6c7eecbd6","schema_version":"1.0","event_id":"sha256:f63f4c04ebaabc35de376cf12194aac757bda0279d4cc7b9bcb94ff6c7eecbd6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:LALRMIHLSIBTET4JWYJEJ3IC7U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MM-MATH: Advancing Multimodal Math Evaluation with Process Evaluation and Fine-grained Classification","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ji Qi, Juanzi Li, Kai Sun, Lei Hou, Yushi Bai","submitted_at":"2024-04-07T22:16:50Z","abstract_excerpt":"To advance the evaluation of multimodal math reasoning in large multimodal models (LMMs), this paper introduces a novel benchmark, MM-MATH. MM-MATH consists of 5,929 open-ended middle school math problems with visual contexts, with fine-grained classification across difficulty, grade level, and knowledge points. Unlike existing benchmarks relying on binary answer comparison, MM-MATH incorporates both outcome and process evaluations. Process evaluation employs LMM-as-a-judge to automatically analyze solution steps, identifying and categorizing errors into specific error types. Extensive evaluat"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.05091","kind":"arxiv","version":4},"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/2404.05091/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:38:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"p6DbQAOft7V7WKRRHYTYoW6tn8Qj9Wh86seJjaooHLx5nmUtf7IO4jPZUIQxSCKCmwhS2XssX0khpKPTcKcoDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T04:35:48.527747Z"},"content_sha256":"5ee9d29a09691db9b0b1d1a3543c5a56e2fd36ef8a2a2a522040568644520fe5","schema_version":"1.0","event_id":"sha256:5ee9d29a09691db9b0b1d1a3543c5a56e2fd36ef8a2a2a522040568644520fe5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LALRMIHLSIBTET4JWYJEJ3IC7U/bundle.json","state_url":"https://pith.science/pith/LALRMIHLSIBTET4JWYJEJ3IC7U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LALRMIHLSIBTET4JWYJEJ3IC7U/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-07T04:35:48Z","links":{"resolver":"https://pith.science/pith/LALRMIHLSIBTET4JWYJEJ3IC7U","bundle":"https://pith.science/pith/LALRMIHLSIBTET4JWYJEJ3IC7U/bundle.json","state":"https://pith.science/pith/LALRMIHLSIBTET4JWYJEJ3IC7U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LALRMIHLSIBTET4JWYJEJ3IC7U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:LALRMIHLSIBTET4JWYJEJ3IC7U","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":"5505e39cdf99c7d2a040a8573422e1df08ae36d96a7b78ff7013284008bf50af","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-07T22:16:50Z","title_canon_sha256":"942e3acb4b66f699034e5bc3b550699a3256d63ae89f2f1d550cbcc984d38d81"},"schema_version":"1.0","source":{"id":"2404.05091","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.05091","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"arxiv_version","alias_value":"2404.05091v4","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.05091","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"pith_short_12","alias_value":"LALRMIHLSIBT","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"pith_short_16","alias_value":"LALRMIHLSIBTET4J","created_at":"2026-07-05T08:38:58Z"},{"alias_kind":"pith_short_8","alias_value":"LALRMIHL","created_at":"2026-07-05T08:38:58Z"}],"graph_snapshots":[{"event_id":"sha256:5ee9d29a09691db9b0b1d1a3543c5a56e2fd36ef8a2a2a522040568644520fe5","target":"graph","created_at":"2026-07-05T08:38:58Z","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/2404.05091/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"To advance the evaluation of multimodal math reasoning in large multimodal models (LMMs), this paper introduces a novel benchmark, MM-MATH. MM-MATH consists of 5,929 open-ended middle school math problems with visual contexts, with fine-grained classification across difficulty, grade level, and knowledge points. Unlike existing benchmarks relying on binary answer comparison, MM-MATH incorporates both outcome and process evaluations. Process evaluation employs LMM-as-a-judge to automatically analyze solution steps, identifying and categorizing errors into specific error types. Extensive evaluat","authors_text":"Ji Qi, Juanzi Li, Kai Sun, Lei Hou, Yushi Bai","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-07T22:16:50Z","title":"MM-MATH: Advancing Multimodal Math Evaluation with Process Evaluation and Fine-grained Classification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.05091","kind":"arxiv","version":4},"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:f63f4c04ebaabc35de376cf12194aac757bda0279d4cc7b9bcb94ff6c7eecbd6","target":"record","created_at":"2026-07-05T08:38:58Z","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":"5505e39cdf99c7d2a040a8573422e1df08ae36d96a7b78ff7013284008bf50af","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-07T22:16:50Z","title_canon_sha256":"942e3acb4b66f699034e5bc3b550699a3256d63ae89f2f1d550cbcc984d38d81"},"schema_version":"1.0","source":{"id":"2404.05091","kind":"arxiv","version":4}},"canonical_sha256":"58171620eb9203324f89b61244ed02fd2a9ded58b04d7f3e2bf8d985577311f7","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"58171620eb9203324f89b61244ed02fd2a9ded58b04d7f3e2bf8d985577311f7","first_computed_at":"2026-07-05T08:38:58.958390Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:38:58.958390Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hvSO62Kd0UfO4fU3/xDYiKuDgMYs/DrptxazYsQdqprbHF12t67VXqKBYGtk7jBP7tAgqakkosVQbo8oCoxICg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:38:58.958870Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.05091","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f63f4c04ebaabc35de376cf12194aac757bda0279d4cc7b9bcb94ff6c7eecbd6","sha256:5ee9d29a09691db9b0b1d1a3543c5a56e2fd36ef8a2a2a522040568644520fe5"],"state_sha256":"adcefe3638ab609edb0c606fab28f58051e61067b6c88d9e5fef1a5c10a7bb5e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lkwKciK6H3Uk2qAhAUyp25JkFr8O+djavxakHMIqUOdy35Lrd08TPiJqi0K5147ZuikjgQElbgPy7od/UgnyCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T04:35:48.529905Z","bundle_sha256":"007e2a8373ccd21ac62dbc7a99cbe37178f84e07c4394d2cde06b2ae434294d6"}}