{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ","short_pith_number":"pith:7OHBPFXN","canonical_record":{"source":{"id":"2506.22992","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-28T19:44:32Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"2771c4b025ec2d5a2eee375b68b9784c5c36f1227993b28f66d8cad904fdb283","abstract_canon_sha256":"2b6de307da51b3ff7312c3ea26214bbed2a5c69f44c5e4c7c6a4ddb97da651cb"},"schema_version":"1.0"},"canonical_sha256":"fb8e1796ed31f030e74154a474f729441befea1d758425204535d13490045969","source":{"kind":"arxiv","id":"2506.22992","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.22992","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"arxiv_version","alias_value":"2506.22992v1","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.22992","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"pith_short_12","alias_value":"7OHBPFXNGHYD","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"pith_short_16","alias_value":"7OHBPFXNGHYDBZ2B","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"pith_short_8","alias_value":"7OHBPFXN","created_at":"2026-07-05T11:28:48Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ","target":"record","payload":{"canonical_record":{"source":{"id":"2506.22992","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-28T19:44:32Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"2771c4b025ec2d5a2eee375b68b9784c5c36f1227993b28f66d8cad904fdb283","abstract_canon_sha256":"2b6de307da51b3ff7312c3ea26214bbed2a5c69f44c5e4c7c6a4ddb97da651cb"},"schema_version":"1.0"},"canonical_sha256":"fb8e1796ed31f030e74154a474f729441befea1d758425204535d13490045969","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:28:48.900391Z","signature_b64":"9zrLcyxeHWuijjeOZBUjG3Z8GXaRYy2XJC3qWkbMvrZduWEmg6AMeCRsm2S9eAs6bhcH27X6wLiFjgZjvJYzAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fb8e1796ed31f030e74154a474f729441befea1d758425204535d13490045969","last_reissued_at":"2026-07-05T11:28:48.899896Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:28:48.899896Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.22992","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:28:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GqjULv2DolviFB//u+9TI5MPyo17uMGfwmUQMkxnDcjnVWGKS7W1iI7DtUlodr9hgUNoTrcMuzGjwCzUDYHcBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:19:02.823799Z"},"content_sha256":"c5496d2d3e2196da0cb6b7539b9fb375d927b609febed7ddbadcc5db71f95c5b","schema_version":"1.0","event_id":"sha256:c5496d2d3e2196da0cb6b7539b9fb375d927b609febed7ddbadcc5db71f95c5b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MARBLE: A Hard Benchmark for Multimodal Spatial Reasoning and Planning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.AI","authors_text":"Maria Brbi\\'c, Michael Moor, Yekun Chai, Yulun Jiang","submitted_at":"2025-06-28T19:44:32Z","abstract_excerpt":"The ability to process information from multiple modalities and to reason through it step-by-step remains a critical challenge in advancing artificial intelligence. However, existing reasoning benchmarks focus on text-only reasoning, or employ multimodal questions that can be answered by directly retrieving information from a non-text modality. Thus, complex reasoning remains poorly understood in multimodal domains. Here, we present MARBLE, a challenging multimodal reasoning benchmark that is designed to scrutinize multimodal language models (MLLMs) in their ability to carefully reason step-by"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.22992","kind":"arxiv","version":1},"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/2506.22992/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":"2026-07-05T11:28:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"omG2L3o3xUA8mllBVp6MVvoxFQCvPDbAAY5qce3h5WSeKanO9UgKpJ7arSYZuivCVHCOOrklloclIkcXEp91BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T12:19:02.824171Z"},"content_sha256":"48ce7b738cbf4ae65a2ac07c743b25a74dcab5adfbc80827659f2e6f936d357d","schema_version":"1.0","event_id":"sha256:48ce7b738cbf4ae65a2ac07c743b25a74dcab5adfbc80827659f2e6f936d357d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ/bundle.json","state_url":"https://pith.science/pith/7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ/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":"2026-07-07T12:19:02Z","links":{"resolver":"https://pith.science/pith/7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ","bundle":"https://pith.science/pith/7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ/bundle.json","state":"https://pith.science/pith/7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7OHBPFXNGHYDBZ2BKSSHJ5ZJIQ","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":"2b6de307da51b3ff7312c3ea26214bbed2a5c69f44c5e4c7c6a4ddb97da651cb","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-28T19:44:32Z","title_canon_sha256":"2771c4b025ec2d5a2eee375b68b9784c5c36f1227993b28f66d8cad904fdb283"},"schema_version":"1.0","source":{"id":"2506.22992","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.22992","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"arxiv_version","alias_value":"2506.22992v1","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.22992","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"pith_short_12","alias_value":"7OHBPFXNGHYD","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"pith_short_16","alias_value":"7OHBPFXNGHYDBZ2B","created_at":"2026-07-05T11:28:48Z"},{"alias_kind":"pith_short_8","alias_value":"7OHBPFXN","created_at":"2026-07-05T11:28:48Z"}],"graph_snapshots":[{"event_id":"sha256:48ce7b738cbf4ae65a2ac07c743b25a74dcab5adfbc80827659f2e6f936d357d","target":"graph","created_at":"2026-07-05T11:28:48Z","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/2506.22992/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The ability to process information from multiple modalities and to reason through it step-by-step remains a critical challenge in advancing artificial intelligence. However, existing reasoning benchmarks focus on text-only reasoning, or employ multimodal questions that can be answered by directly retrieving information from a non-text modality. Thus, complex reasoning remains poorly understood in multimodal domains. Here, we present MARBLE, a challenging multimodal reasoning benchmark that is designed to scrutinize multimodal language models (MLLMs) in their ability to carefully reason step-by","authors_text":"Maria Brbi\\'c, Michael Moor, Yekun Chai, Yulun Jiang","cross_cats":["cs.CL","cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-28T19:44:32Z","title":"MARBLE: A Hard Benchmark for Multimodal Spatial Reasoning and Planning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.22992","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:c5496d2d3e2196da0cb6b7539b9fb375d927b609febed7ddbadcc5db71f95c5b","target":"record","created_at":"2026-07-05T11:28:48Z","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":"2b6de307da51b3ff7312c3ea26214bbed2a5c69f44c5e4c7c6a4ddb97da651cb","cross_cats_sorted":["cs.CL","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-28T19:44:32Z","title_canon_sha256":"2771c4b025ec2d5a2eee375b68b9784c5c36f1227993b28f66d8cad904fdb283"},"schema_version":"1.0","source":{"id":"2506.22992","kind":"arxiv","version":1}},"canonical_sha256":"fb8e1796ed31f030e74154a474f729441befea1d758425204535d13490045969","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"fb8e1796ed31f030e74154a474f729441befea1d758425204535d13490045969","first_computed_at":"2026-07-05T11:28:48.899896Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:28:48.899896Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9zrLcyxeHWuijjeOZBUjG3Z8GXaRYy2XJC3qWkbMvrZduWEmg6AMeCRsm2S9eAs6bhcH27X6wLiFjgZjvJYzAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:28:48.900391Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.22992","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c5496d2d3e2196da0cb6b7539b9fb375d927b609febed7ddbadcc5db71f95c5b","sha256:48ce7b738cbf4ae65a2ac07c743b25a74dcab5adfbc80827659f2e6f936d357d"],"state_sha256":"ff6ac96f3c6549d7d66198dc36c647ab2e08c475650257902053da523e8aa891"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NMJ/yH9tUk1VxgvvBIkEUpebwIPKacmtnad4jVuPHPgBr+fAI+giixwQGNWl8ZVP/Gp9rPVQdDu6SYOYtyRUCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T12:19:02.826156Z","bundle_sha256":"e4daf633606bd7391c4e3578a8c3156369216ea756f7bb66d91814855ec750f8"}}