{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:COFR6SFESUMGGCYAHJHJ2O2OAR","short_pith_number":"pith:COFR6SFE","canonical_record":{"source":{"id":"2410.13754","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-17T16:52:28Z","cross_cats_sorted":["cs.LG","cs.MM"],"title_canon_sha256":"fe5128b3b59c64c1465d608e338b23066afd4f6d21a1df40dad8193a89a936f7","abstract_canon_sha256":"db190a0f859657af6eb920af19bd42de2303ecd1845d315fadbb01011c1d33c8"},"schema_version":"1.0"},"canonical_sha256":"138b1f48a49518630b003a4e9d3b4e0460f76b32fdb7f3cffe90d6ce1dc206bd","source":{"kind":"arxiv","id":"2410.13754","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.13754","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"arxiv_version","alias_value":"2410.13754v2","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.13754","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"pith_short_12","alias_value":"COFR6SFESUMG","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"pith_short_16","alias_value":"COFR6SFESUMGGCYA","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"pith_short_8","alias_value":"COFR6SFE","created_at":"2026-07-05T09:22:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:COFR6SFESUMGGCYAHJHJ2O2OAR","target":"record","payload":{"canonical_record":{"source":{"id":"2410.13754","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-17T16:52:28Z","cross_cats_sorted":["cs.LG","cs.MM"],"title_canon_sha256":"fe5128b3b59c64c1465d608e338b23066afd4f6d21a1df40dad8193a89a936f7","abstract_canon_sha256":"db190a0f859657af6eb920af19bd42de2303ecd1845d315fadbb01011c1d33c8"},"schema_version":"1.0"},"canonical_sha256":"138b1f48a49518630b003a4e9d3b4e0460f76b32fdb7f3cffe90d6ce1dc206bd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:22:24.337780Z","signature_b64":"1bBf9QilddEMWKMOfQKqgV/8mSWRABfKFhXB7HxR5xFDDlvebqzAnrpHQYsjfhSADq5spK5ywXqZN0hoFGKGDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"138b1f48a49518630b003a4e9d3b4e0460f76b32fdb7f3cffe90d6ce1dc206bd","last_reissued_at":"2026-07-05T09:22:24.337190Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:22:24.337190Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2410.13754","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:22:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t1tBTWzpG3JmDjDBqP0SXl4Y7GEIhxQwU8sj9+V+8h8u3kWfSEFzEQx2wyKhLUP8TTZ3FT5g3PNZi3HIenGQAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-23T05:02:21.993079Z"},"content_sha256":"20590bdfc38a50829cf03c50dfcf0f9f872fd23b8eab732a8f9fa8d36972c819","schema_version":"1.0","event_id":"sha256:20590bdfc38a50829cf03c50dfcf0f9f872fd23b8eab732a8f9fa8d36972c819"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:COFR6SFESUMGGCYAHJHJ2O2OAR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MixEval-X: Any-to-Any Evaluations from Real-World Data Mixtures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.MM"],"primary_cat":"cs.AI","authors_text":"Bo Li, David Junhao Zhang, Deepanway Ghosal, Fuzhao Xue, Jinjie Ni, Kabir Jain, Kaichen Zhang, Mahir Shah, Michael Shieh, Xiang Yue, Yang You, Yifan Song, Zian Zheng","submitted_at":"2024-10-17T16:52:28Z","abstract_excerpt":"Perceiving and generating diverse modalities are crucial for AI models to effectively learn from and engage with real-world signals, necessitating reliable evaluations for their development. We identify two major issues in current evaluations: (1) inconsistent standards, shaped by different communities with varying protocols and maturity levels; and (2) significant query, grading, and generalization biases. To address these, we introduce MixEval-X, the first any-to-any, real-world benchmark designed to optimize and standardize evaluations across diverse input and output modalities. We propose "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.13754","kind":"arxiv","version":2},"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/2410.13754/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":"2026-07-05T09:22:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qJnYyAHmyXLFtc8F0GuK8f6eZR6B+curGfy5G3eGJOTduUElYFjlSLew6qhd4JipNxW/hmAihy1f4CH95KArDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-23T05:02:21.993479Z"},"content_sha256":"c733aaebe0f6ed33b03e25734180dab8181e3334413932da92f0b70efa5bdf55","schema_version":"1.0","event_id":"sha256:c733aaebe0f6ed33b03e25734180dab8181e3334413932da92f0b70efa5bdf55"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/COFR6SFESUMGGCYAHJHJ2O2OAR/bundle.json","state_url":"https://pith.science/pith/COFR6SFESUMGGCYAHJHJ2O2OAR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/COFR6SFESUMGGCYAHJHJ2O2OAR/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":"2026-07-23T05:02:21Z","links":{"resolver":"https://pith.science/pith/COFR6SFESUMGGCYAHJHJ2O2OAR","bundle":"https://pith.science/pith/COFR6SFESUMGGCYAHJHJ2O2OAR/bundle.json","state":"https://pith.science/pith/COFR6SFESUMGGCYAHJHJ2O2OAR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/COFR6SFESUMGGCYAHJHJ2O2OAR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:COFR6SFESUMGGCYAHJHJ2O2OAR","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":"db190a0f859657af6eb920af19bd42de2303ecd1845d315fadbb01011c1d33c8","cross_cats_sorted":["cs.LG","cs.MM"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-17T16:52:28Z","title_canon_sha256":"fe5128b3b59c64c1465d608e338b23066afd4f6d21a1df40dad8193a89a936f7"},"schema_version":"1.0","source":{"id":"2410.13754","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2410.13754","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"arxiv_version","alias_value":"2410.13754v2","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.13754","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"pith_short_12","alias_value":"COFR6SFESUMG","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"pith_short_16","alias_value":"COFR6SFESUMGGCYA","created_at":"2026-07-05T09:22:24Z"},{"alias_kind":"pith_short_8","alias_value":"COFR6SFE","created_at":"2026-07-05T09:22:24Z"}],"graph_snapshots":[{"event_id":"sha256:c733aaebe0f6ed33b03e25734180dab8181e3334413932da92f0b70efa5bdf55","target":"graph","created_at":"2026-07-05T09:22:24Z","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/2410.13754/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Perceiving and generating diverse modalities are crucial for AI models to effectively learn from and engage with real-world signals, necessitating reliable evaluations for their development. We identify two major issues in current evaluations: (1) inconsistent standards, shaped by different communities with varying protocols and maturity levels; and (2) significant query, grading, and generalization biases. To address these, we introduce MixEval-X, the first any-to-any, real-world benchmark designed to optimize and standardize evaluations across diverse input and output modalities. We propose ","authors_text":"Bo Li, David Junhao Zhang, Deepanway Ghosal, Fuzhao Xue, Jinjie Ni, Kabir Jain, Kaichen Zhang, Mahir Shah, Michael Shieh, Xiang Yue, Yang You, Yifan Song, Zian Zheng","cross_cats":["cs.LG","cs.MM"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-17T16:52:28Z","title":"MixEval-X: Any-to-Any Evaluations from Real-World Data Mixtures"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.13754","kind":"arxiv","version":2},"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":"sha256:20590bdfc38a50829cf03c50dfcf0f9f872fd23b8eab732a8f9fa8d36972c819","target":"record","created_at":"2026-07-05T09:22:24Z","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":"db190a0f859657af6eb920af19bd42de2303ecd1845d315fadbb01011c1d33c8","cross_cats_sorted":["cs.LG","cs.MM"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-10-17T16:52:28Z","title_canon_sha256":"fe5128b3b59c64c1465d608e338b23066afd4f6d21a1df40dad8193a89a936f7"},"schema_version":"1.0","source":{"id":"2410.13754","kind":"arxiv","version":2}},"canonical_sha256":"138b1f48a49518630b003a4e9d3b4e0460f76b32fdb7f3cffe90d6ce1dc206bd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"138b1f48a49518630b003a4e9d3b4e0460f76b32fdb7f3cffe90d6ce1dc206bd","first_computed_at":"2026-07-05T09:22:24.337190Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:22:24.337190Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1bBf9QilddEMWKMOfQKqgV/8mSWRABfKFhXB7HxR5xFDDlvebqzAnrpHQYsjfhSADq5spK5ywXqZN0hoFGKGDA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:22:24.337780Z","signed_message":"canonical_sha256_bytes"},"source_id":"2410.13754","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:20590bdfc38a50829cf03c50dfcf0f9f872fd23b8eab732a8f9fa8d36972c819","sha256:c733aaebe0f6ed33b03e25734180dab8181e3334413932da92f0b70efa5bdf55"],"state_sha256":"a42f4780d32997ea3b54b0dae10fb9fdaefad6fb18f7bcf5672520dd225c2bcc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PurgT/bEo87zPtMvTHGENNeAAr5iowO+OKi/WVU0tUC03gttHapRKqzXhJXUpwx7W9sJ8KTJpDUC7ZdzpYfoCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-23T05:02:21.995735Z","bundle_sha256":"1095ea3cd20284251ffad0463941c93aa9d983a390f311730a0b9dab9829429d"}}