{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:KR2MGAZP75F3RQM3LSU4KGNE4H","short_pith_number":"pith:KR2MGAZP","canonical_record":{"source":{"id":"2403.03864","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-03-06T17:15:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"69b30c9937422081616230ff49b1b198e4eac82c05817683b9deb1208b30f320","abstract_canon_sha256":"63ce2832adce5607242191ce5e92d4789bb185a7c6046a3679d0959df7499255"},"schema_version":"1.0"},"canonical_sha256":"5474c3032fff4bb8c19b5ca9c519a4e1e33f61ba93388ddd3c06c83cb321b19e","source":{"kind":"arxiv","id":"2403.03864","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.03864","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"arxiv_version","alias_value":"2403.03864v3","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.03864","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"pith_short_12","alias_value":"KR2MGAZP75F3","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"pith_short_16","alias_value":"KR2MGAZP75F3RQM3","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"pith_short_8","alias_value":"KR2MGAZP","created_at":"2026-07-05T07:55:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:KR2MGAZP75F3RQM3LSU4KGNE4H","target":"record","payload":{"canonical_record":{"source":{"id":"2403.03864","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-03-06T17:15:04Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"69b30c9937422081616230ff49b1b198e4eac82c05817683b9deb1208b30f320","abstract_canon_sha256":"63ce2832adce5607242191ce5e92d4789bb185a7c6046a3679d0959df7499255"},"schema_version":"1.0"},"canonical_sha256":"5474c3032fff4bb8c19b5ca9c519a4e1e33f61ba93388ddd3c06c83cb321b19e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:55:29.268375Z","signature_b64":"ZUUajBRrGokDiKvrzuDeUK7wvdUIfnOh+vg58i6Z+bfozRA29Vrd6HaO5775RjCKMBCzzFoLBDZ8O2lYm3BpDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5474c3032fff4bb8c19b5ca9c519a4e1e33f61ba93388ddd3c06c83cb321b19e","last_reissued_at":"2026-07-05T07:55:29.267915Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:55:29.267915Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.03864","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:55:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TswNt5+Qot+dRbcyCbiWOimPdrW58+DkNwd/BzrdbIIl5z6zzRREobsGdNs7jQsq9MDmiKDz4UPbmg8/Ft8PCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T09:36:09.041717Z"},"content_sha256":"f4b3906e24c8bf9c76cbfa11099cb77d77b707ea88603030ef892823c4f09fc9","schema_version":"1.0","event_id":"sha256:f4b3906e24c8bf9c76cbfa11099cb77d77b707ea88603030ef892823c4f09fc9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:KR2MGAZP75F3RQM3LSU4KGNE4H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Are Language Models Puzzle Prodigies? Algorithmic Puzzles Unveil Serious Challenges in Multimodal Reasoning","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Chia Yew Ken, Deepanway Ghosal, Soujanya Poria, Vernon Toh Yan Han","submitted_at":"2024-03-06T17:15:04Z","abstract_excerpt":"This paper introduces the novel task of multimodal puzzle solving, framed within the context of visual question-answering. We present a new dataset, AlgoPuzzleVQA designed to challenge and evaluate the capabilities of multimodal language models in solving algorithmic puzzles that necessitate both visual understanding, language understanding, and complex algorithmic reasoning. We create the puzzles to encompass a diverse array of mathematical and algorithmic topics such as boolean logic, combinatorics, graph theory, optimization, search, etc., aiming to evaluate the gap between visual data inte"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.03864","kind":"arxiv","version":3},"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/2403.03864/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":"2026-07-05T07:55:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LF0vFDrN+jRKzAgffau4mdKuSO5SeoGEkmApsJr9WRWh3ee3W+DVdaF1/72f2D53vSjo49JWJst8hHut8O9qAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-12T09:36:09.042094Z"},"content_sha256":"302445d1ef82fc098f619a6400c74c204e4008cd8ced00bea35c749bc21668a1","schema_version":"1.0","event_id":"sha256:302445d1ef82fc098f619a6400c74c204e4008cd8ced00bea35c749bc21668a1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KR2MGAZP75F3RQM3LSU4KGNE4H/bundle.json","state_url":"https://pith.science/pith/KR2MGAZP75F3RQM3LSU4KGNE4H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KR2MGAZP75F3RQM3LSU4KGNE4H/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":"2026-07-12T09:36:09Z","links":{"resolver":"https://pith.science/pith/KR2MGAZP75F3RQM3LSU4KGNE4H","bundle":"https://pith.science/pith/KR2MGAZP75F3RQM3LSU4KGNE4H/bundle.json","state":"https://pith.science/pith/KR2MGAZP75F3RQM3LSU4KGNE4H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KR2MGAZP75F3RQM3LSU4KGNE4H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:KR2MGAZP75F3RQM3LSU4KGNE4H","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":"63ce2832adce5607242191ce5e92d4789bb185a7c6046a3679d0959df7499255","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-03-06T17:15:04Z","title_canon_sha256":"69b30c9937422081616230ff49b1b198e4eac82c05817683b9deb1208b30f320"},"schema_version":"1.0","source":{"id":"2403.03864","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.03864","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"arxiv_version","alias_value":"2403.03864v3","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.03864","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"pith_short_12","alias_value":"KR2MGAZP75F3","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"pith_short_16","alias_value":"KR2MGAZP75F3RQM3","created_at":"2026-07-05T07:55:29Z"},{"alias_kind":"pith_short_8","alias_value":"KR2MGAZP","created_at":"2026-07-05T07:55:29Z"}],"graph_snapshots":[{"event_id":"sha256:302445d1ef82fc098f619a6400c74c204e4008cd8ced00bea35c749bc21668a1","target":"graph","created_at":"2026-07-05T07:55:29Z","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/2403.03864/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper introduces the novel task of multimodal puzzle solving, framed within the context of visual question-answering. We present a new dataset, AlgoPuzzleVQA designed to challenge and evaluate the capabilities of multimodal language models in solving algorithmic puzzles that necessitate both visual understanding, language understanding, and complex algorithmic reasoning. We create the puzzles to encompass a diverse array of mathematical and algorithmic topics such as boolean logic, combinatorics, graph theory, optimization, search, etc., aiming to evaluate the gap between visual data inte","authors_text":"Chia Yew Ken, Deepanway Ghosal, Soujanya Poria, Vernon Toh Yan Han","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-03-06T17:15:04Z","title":"Are Language Models Puzzle Prodigies? Algorithmic Puzzles Unveil Serious Challenges in Multimodal Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.03864","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:f4b3906e24c8bf9c76cbfa11099cb77d77b707ea88603030ef892823c4f09fc9","target":"record","created_at":"2026-07-05T07:55:29Z","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":"63ce2832adce5607242191ce5e92d4789bb185a7c6046a3679d0959df7499255","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2024-03-06T17:15:04Z","title_canon_sha256":"69b30c9937422081616230ff49b1b198e4eac82c05817683b9deb1208b30f320"},"schema_version":"1.0","source":{"id":"2403.03864","kind":"arxiv","version":3}},"canonical_sha256":"5474c3032fff4bb8c19b5ca9c519a4e1e33f61ba93388ddd3c06c83cb321b19e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5474c3032fff4bb8c19b5ca9c519a4e1e33f61ba93388ddd3c06c83cb321b19e","first_computed_at":"2026-07-05T07:55:29.267915Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:55:29.267915Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ZUUajBRrGokDiKvrzuDeUK7wvdUIfnOh+vg58i6Z+bfozRA29Vrd6HaO5775RjCKMBCzzFoLBDZ8O2lYm3BpDw==","signature_status":"signed_v1","signed_at":"2026-07-05T07:55:29.268375Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.03864","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f4b3906e24c8bf9c76cbfa11099cb77d77b707ea88603030ef892823c4f09fc9","sha256:302445d1ef82fc098f619a6400c74c204e4008cd8ced00bea35c749bc21668a1"],"state_sha256":"ab77c30b84e59d033b4a0758131e672367a4fc3af1da20c5979e971892874bd8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VYSsvCQ2Hf154HC8QfBJzsTWLd3WSA3/Fb0qHUBHSLhguTU1sbsW4mEksLinEIsCiVnkT0nCCrpFwl1Lal5fDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-12T09:36:09.045344Z","bundle_sha256":"1dd13840c86d6cb757cef409b7154e3d3e8a0236cf8c78983432817958a9100d"}}