{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:OUGASVG4UW2FVSYNNYQDQ7DTAK","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":"58d75c66c58221bf28de21d6bb9dd805a612713e094156c9d1e4a4098e59b761","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-04-01T17:43:27Z","title_canon_sha256":"15d54361fafd22db9fa80c16b012975072818f195b5c716af8e96d040ba4470b"},"schema_version":"1.0","source":{"id":"2404.01266","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.01266","created_at":"2026-07-05T08:56:18Z"},{"alias_kind":"arxiv_version","alias_value":"2404.01266v3","created_at":"2026-07-05T08:56:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.01266","created_at":"2026-07-05T08:56:18Z"},{"alias_kind":"pith_short_12","alias_value":"OUGASVG4UW2F","created_at":"2026-07-05T08:56:18Z"},{"alias_kind":"pith_short_16","alias_value":"OUGASVG4UW2FVSYN","created_at":"2026-07-05T08:56:18Z"},{"alias_kind":"pith_short_8","alias_value":"OUGASVG4","created_at":"2026-07-05T08:56:18Z"}],"graph_snapshots":[{"event_id":"sha256:af6bc8b0c81dce9b35b08781067bfcd63ecc10979603bedd1a5b721a40fe2116","target":"graph","created_at":"2026-07-05T08:56:18Z","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/2404.01266/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current foundation models exhibit impressive capabilities when prompted either with text only or with both image and text inputs. But do their capabilities change depending on the input modality? In this work, we propose $\\textbf{IsoBench}$, a benchmark dataset containing problems from four major areas: math, science, algorithms, and games. Each example is presented with multiple $\\textbf{isomorphic representations}$ of inputs, such as visual, textual, and mathematical presentations. IsoBench provides fine-grained feedback to diagnose performance gaps caused by the form of the representation. ","authors_text":"Bhuwan Dhingra, Dani Yogatama, Deqing Fu, Ghazal Khalighinejad, Ollie Liu, Robin Jia, Ruohao Guo, Willie Neiswanger","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-04-01T17:43:27Z","title":"IsoBench: Benchmarking Multimodal Foundation Models on Isomorphic Representations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.01266","kind":"arxiv","version":3},"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":"sha256:6059cdd25881ff51f15b4aef14d86001d07b3d87bd654ab7c6ef9db5de742ae7","target":"record","created_at":"2026-07-05T08:56:18Z","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":"58d75c66c58221bf28de21d6bb9dd805a612713e094156c9d1e4a4098e59b761","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-04-01T17:43:27Z","title_canon_sha256":"15d54361fafd22db9fa80c16b012975072818f195b5c716af8e96d040ba4470b"},"schema_version":"1.0","source":{"id":"2404.01266","kind":"arxiv","version":3}},"canonical_sha256":"750c0954dca5b45acb0d6e20387c7302bcb6891d0781104bfeb26074358b9dfe","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"750c0954dca5b45acb0d6e20387c7302bcb6891d0781104bfeb26074358b9dfe","first_computed_at":"2026-07-05T08:56:18.395739Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:56:18.395739Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3JWUtYTDts+Oo5drd4PR+xEB4dT6x2cE3svIDRQBDcsPwpuQZOtxkce1Soj1vLDyqsXsjWSy5mM5MZaWWhNeCg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:56:18.396187Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.01266","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6059cdd25881ff51f15b4aef14d86001d07b3d87bd654ab7c6ef9db5de742ae7","sha256:af6bc8b0c81dce9b35b08781067bfcd63ecc10979603bedd1a5b721a40fe2116"],"state_sha256":"7867aae1109606865a71b40371483d5e39c68cc2a3aaebc1258e23794d5b0c89"}