{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:LATZJ53XCPJ2KHGA7HJTCTTHSW","short_pith_number":"pith:LATZJ53X","canonical_record":{"source":{"id":"2505.00681","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-01T17:41:49Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"b48c5a4409abc5e814f3a20133606c76ecb9a7a8980be72dc40a6046fd7a0b54","abstract_canon_sha256":"a7ee7976855ddb53d7645e12c8880ab782d712e7f788b7137c89e73bf86a1087"},"schema_version":"1.0"},"canonical_sha256":"582794f77713d3a51cc0f9d3314e6795989ed580d22234b5daa52983b09be111","source":{"kind":"arxiv","id":"2505.00681","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.00681","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"arxiv_version","alias_value":"2505.00681v1","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.00681","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"pith_short_12","alias_value":"LATZJ53XCPJ2","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"pith_short_16","alias_value":"LATZJ53XCPJ2KHGA","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"pith_short_8","alias_value":"LATZJ53X","created_at":"2026-07-05T10:57:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:LATZJ53XCPJ2KHGA7HJTCTTHSW","target":"record","payload":{"canonical_record":{"source":{"id":"2505.00681","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-01T17:41:49Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"b48c5a4409abc5e814f3a20133606c76ecb9a7a8980be72dc40a6046fd7a0b54","abstract_canon_sha256":"a7ee7976855ddb53d7645e12c8880ab782d712e7f788b7137c89e73bf86a1087"},"schema_version":"1.0"},"canonical_sha256":"582794f77713d3a51cc0f9d3314e6795989ed580d22234b5daa52983b09be111","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:57:24.016122Z","signature_b64":"5hjdUVBQHTqeJg2DjMmMru0/+PPy144qQjpVAwpfsyyxdIs/XYVkSY0ujma9uDKrYMvnEv30150xBhhHBHV6CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"582794f77713d3a51cc0f9d3314e6795989ed580d22234b5daa52983b09be111","last_reissued_at":"2026-07-05T10:57:24.015651Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:57:24.015651Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.00681","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":"2026-07-05T10:57:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9zlk0iN3eU+YhlcdxOZ22jqTFRzbetsBwQJ1pd0t7zoP554IgW/QYyMAlrLfIrA/Gj1vplLOSvvwcA9LhihWBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T06:50:55.789545Z"},"content_sha256":"df0557a001e2d640631d56cb8c0debbbb4b2ad4e12769967f616c14ea2c16943","schema_version":"1.0","event_id":"sha256:df0557a001e2d640631d56cb8c0debbbb4b2ad4e12769967f616c14ea2c16943"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:LATZJ53XCPJ2KHGA7HJTCTTHSW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"MINERVA: Evaluating Complex Video Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Ahmet Iscen, Anja Hauth, Arsha Nagrani, Carl Vondrick, Cordelia Schmid, Mikhail Sirotenko, Nilpa Jha, Ramin Mehran, Sachit Menon, Shyamal Buch, Tobias Weyand, Yukun Zhu","submitted_at":"2025-05-01T17:41:49Z","abstract_excerpt":"Multimodal LLMs are turning their focus to video benchmarks, however most video benchmarks only provide outcome supervision, with no intermediate or interpretable reasoning steps. This makes it challenging to assess if models are truly able to combine perceptual and temporal information to reason about videos, or simply get the correct answer by chance or by exploiting linguistic biases. To remedy this, we provide a new video reasoning dataset called MINERVA for modern multimodal models. Each question in the dataset comes with 5 answer choices, as well as detailed, hand-crafted reasoning trace"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.00681","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/2505.00681/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":"2026-07-05T10:57:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"chrWXiShvMiHHN2E+sWlUOd/m/qaA6NdzZeudAIGTX1D1l4btnuouh6xTzcCV91IpCuUWuhl7lxdqM5aMVRRCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T06:50:55.790043Z"},"content_sha256":"583141f276df2ccec0818e4e741db86e0760c82b2310a440ede945d48ad116da","schema_version":"1.0","event_id":"sha256:583141f276df2ccec0818e4e741db86e0760c82b2310a440ede945d48ad116da"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LATZJ53XCPJ2KHGA7HJTCTTHSW/bundle.json","state_url":"https://pith.science/pith/LATZJ53XCPJ2KHGA7HJTCTTHSW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LATZJ53XCPJ2KHGA7HJTCTTHSW/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":"2026-08-07T06:50:55Z","links":{"resolver":"https://pith.science/pith/LATZJ53XCPJ2KHGA7HJTCTTHSW","bundle":"https://pith.science/pith/LATZJ53XCPJ2KHGA7HJTCTTHSW/bundle.json","state":"https://pith.science/pith/LATZJ53XCPJ2KHGA7HJTCTTHSW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LATZJ53XCPJ2KHGA7HJTCTTHSW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:LATZJ53XCPJ2KHGA7HJTCTTHSW","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":"a7ee7976855ddb53d7645e12c8880ab782d712e7f788b7137c89e73bf86a1087","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-01T17:41:49Z","title_canon_sha256":"b48c5a4409abc5e814f3a20133606c76ecb9a7a8980be72dc40a6046fd7a0b54"},"schema_version":"1.0","source":{"id":"2505.00681","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.00681","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"arxiv_version","alias_value":"2505.00681v1","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.00681","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"pith_short_12","alias_value":"LATZJ53XCPJ2","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"pith_short_16","alias_value":"LATZJ53XCPJ2KHGA","created_at":"2026-07-05T10:57:24Z"},{"alias_kind":"pith_short_8","alias_value":"LATZJ53X","created_at":"2026-07-05T10:57:24Z"}],"graph_snapshots":[{"event_id":"sha256:583141f276df2ccec0818e4e741db86e0760c82b2310a440ede945d48ad116da","target":"graph","created_at":"2026-07-05T10:57:24Z","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/2505.00681/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal LLMs are turning their focus to video benchmarks, however most video benchmarks only provide outcome supervision, with no intermediate or interpretable reasoning steps. This makes it challenging to assess if models are truly able to combine perceptual and temporal information to reason about videos, or simply get the correct answer by chance or by exploiting linguistic biases. To remedy this, we provide a new video reasoning dataset called MINERVA for modern multimodal models. Each question in the dataset comes with 5 answer choices, as well as detailed, hand-crafted reasoning trace","authors_text":"Ahmet Iscen, Anja Hauth, Arsha Nagrani, Carl Vondrick, Cordelia Schmid, Mikhail Sirotenko, Nilpa Jha, Ramin Mehran, Sachit Menon, Shyamal Buch, Tobias Weyand, Yukun Zhu","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-01T17:41:49Z","title":"MINERVA: Evaluating Complex Video Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.00681","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:df0557a001e2d640631d56cb8c0debbbb4b2ad4e12769967f616c14ea2c16943","target":"record","created_at":"2026-07-05T10:57:24Z","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":"a7ee7976855ddb53d7645e12c8880ab782d712e7f788b7137c89e73bf86a1087","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-01T17:41:49Z","title_canon_sha256":"b48c5a4409abc5e814f3a20133606c76ecb9a7a8980be72dc40a6046fd7a0b54"},"schema_version":"1.0","source":{"id":"2505.00681","kind":"arxiv","version":1}},"canonical_sha256":"582794f77713d3a51cc0f9d3314e6795989ed580d22234b5daa52983b09be111","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"582794f77713d3a51cc0f9d3314e6795989ed580d22234b5daa52983b09be111","first_computed_at":"2026-07-05T10:57:24.015651Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:57:24.015651Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"5hjdUVBQHTqeJg2DjMmMru0/+PPy144qQjpVAwpfsyyxdIs/XYVkSY0ujma9uDKrYMvnEv30150xBhhHBHV6CQ==","signature_status":"signed_v1","signed_at":"2026-07-05T10:57:24.016122Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.00681","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:df0557a001e2d640631d56cb8c0debbbb4b2ad4e12769967f616c14ea2c16943","sha256:583141f276df2ccec0818e4e741db86e0760c82b2310a440ede945d48ad116da"],"state_sha256":"e1ae82f6747560ca5f690fcf0931eaf2db0cbe686871c07cf1608a27cc278a31"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ycby/Bo1U6JctCzAlQ8Mq7zn89pdnJH7B1inKu9O/dCiQ3BoukXI3Yj3WANLu/piJcQrVe6JC+F/FkLXa6XlDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T06:50:55.795552Z","bundle_sha256":"7ff5bd15baa3e119f1f0776c118bb66e36b67dda666eac660f364743a1dcd96f"}}