{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:QFQWN44TNS5CUJPHFE6PAGGWNM","short_pith_number":"pith:QFQWN44T","canonical_record":{"source":{"id":"2505.14640","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-20T17:26:32Z","cross_cats_sorted":[],"title_canon_sha256":"5022d687ef4d596d506f79047b49527ed014708909a3507703e9bffbbe2e15c9","abstract_canon_sha256":"b8e7585cc723611c02d776132ec6177a1fe812a79e323542a9cfc28f721cbe91"},"schema_version":"1.0"},"canonical_sha256":"816166f3936cba2a25e7293cf018d66b041c13d14c6edcd30bb7686fb18cfb0c","source":{"kind":"arxiv","id":"2505.14640","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.14640","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"arxiv_version","alias_value":"2505.14640v1","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14640","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"pith_short_12","alias_value":"QFQWN44TNS5C","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"pith_short_16","alias_value":"QFQWN44TNS5CUJPH","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"pith_short_8","alias_value":"QFQWN44T","created_at":"2026-07-05T11:06:16Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:QFQWN44TNS5CUJPHFE6PAGGWNM","target":"record","payload":{"canonical_record":{"source":{"id":"2505.14640","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-20T17:26:32Z","cross_cats_sorted":[],"title_canon_sha256":"5022d687ef4d596d506f79047b49527ed014708909a3507703e9bffbbe2e15c9","abstract_canon_sha256":"b8e7585cc723611c02d776132ec6177a1fe812a79e323542a9cfc28f721cbe91"},"schema_version":"1.0"},"canonical_sha256":"816166f3936cba2a25e7293cf018d66b041c13d14c6edcd30bb7686fb18cfb0c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:06:16.894577Z","signature_b64":"qz+dF7lL6995F2SANkn3jb4i4XA3nVcCaH4QOAU8Poq1AJFx4yFxbnCcrxl5D/shG2NaPrb+FPcFp5g81NLXBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"816166f3936cba2a25e7293cf018d66b041c13d14c6edcd30bb7686fb18cfb0c","last_reissued_at":"2026-07-05T11:06:16.894054Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:06:16.894054Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.14640","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":"2026-07-05T11:06:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/sxfPD+jTSfBTteC+PocKUW/KgglEL7pRl/fYETcAvS87P2U+eEn74apCDyHmUinTfBFptfNb7xAuDdb84J8Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T19:21:17.246850Z"},"content_sha256":"8f5cf868d932ad3de1bbde2853f4a747cef7d910bc5008b225a254c236439a15","schema_version":"1.0","event_id":"sha256:8f5cf868d932ad3de1bbde2853f4a747cef7d910bc5008b225a254c236439a15"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:QFQWN44TNS5CUJPHFE6PAGGWNM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"VideoEval-Pro: Robust and Realistic Long Video Understanding Evaluation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Ge Zhang, Ping Nie, Weiming Ren, Wenhu Chen, Wentao Ma, Yiming Jia, Zhuofeng Li","submitted_at":"2025-05-20T17:26:32Z","abstract_excerpt":"Large multimodal models (LMMs) have recently emerged as a powerful tool for long video understanding (LVU), prompting the development of standardized LVU benchmarks to evaluate their performance. However, our investigation reveals a rather sober lesson for existing LVU benchmarks. First, most existing benchmarks rely heavily on multiple-choice questions (MCQs), whose evaluation results are inflated due to the possibility of guessing the correct answer; Second, a significant portion of questions in these benchmarks have strong priors to allow models to answer directly without even reading the i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14640","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/2505.14640/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":"2026-07-05T11:06:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PPtn6RHlFDdXSaujkHjaaHdFgeafHX8GBrlwKvf+Fy8c56h5quuE5BFBY3dRREHBRM91QwMRCH0oHVfGrAnUDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T19:21:17.247967Z"},"content_sha256":"07067a8633c21b301a7e773f104b6f86a34348b4df94fbf1e3c8e54e165f1234","schema_version":"1.0","event_id":"sha256:07067a8633c21b301a7e773f104b6f86a34348b4df94fbf1e3c8e54e165f1234"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QFQWN44TNS5CUJPHFE6PAGGWNM/bundle.json","state_url":"https://pith.science/pith/QFQWN44TNS5CUJPHFE6PAGGWNM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QFQWN44TNS5CUJPHFE6PAGGWNM/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":"2026-08-04T19:21:17Z","links":{"resolver":"https://pith.science/pith/QFQWN44TNS5CUJPHFE6PAGGWNM","bundle":"https://pith.science/pith/QFQWN44TNS5CUJPHFE6PAGGWNM/bundle.json","state":"https://pith.science/pith/QFQWN44TNS5CUJPHFE6PAGGWNM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QFQWN44TNS5CUJPHFE6PAGGWNM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:QFQWN44TNS5CUJPHFE6PAGGWNM","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":"b8e7585cc723611c02d776132ec6177a1fe812a79e323542a9cfc28f721cbe91","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-20T17:26:32Z","title_canon_sha256":"5022d687ef4d596d506f79047b49527ed014708909a3507703e9bffbbe2e15c9"},"schema_version":"1.0","source":{"id":"2505.14640","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.14640","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"arxiv_version","alias_value":"2505.14640v1","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.14640","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"pith_short_12","alias_value":"QFQWN44TNS5C","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"pith_short_16","alias_value":"QFQWN44TNS5CUJPH","created_at":"2026-07-05T11:06:16Z"},{"alias_kind":"pith_short_8","alias_value":"QFQWN44T","created_at":"2026-07-05T11:06:16Z"}],"graph_snapshots":[{"event_id":"sha256:07067a8633c21b301a7e773f104b6f86a34348b4df94fbf1e3c8e54e165f1234","target":"graph","created_at":"2026-07-05T11:06:16Z","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/2505.14640/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large multimodal models (LMMs) have recently emerged as a powerful tool for long video understanding (LVU), prompting the development of standardized LVU benchmarks to evaluate their performance. However, our investigation reveals a rather sober lesson for existing LVU benchmarks. First, most existing benchmarks rely heavily on multiple-choice questions (MCQs), whose evaluation results are inflated due to the possibility of guessing the correct answer; Second, a significant portion of questions in these benchmarks have strong priors to allow models to answer directly without even reading the i","authors_text":"Ge Zhang, Ping Nie, Weiming Ren, Wenhu Chen, Wentao Ma, Yiming Jia, Zhuofeng Li","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-20T17:26:32Z","title":"VideoEval-Pro: Robust and Realistic Long Video Understanding Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.14640","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":"sha256:8f5cf868d932ad3de1bbde2853f4a747cef7d910bc5008b225a254c236439a15","target":"record","created_at":"2026-07-05T11:06:16Z","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":"b8e7585cc723611c02d776132ec6177a1fe812a79e323542a9cfc28f721cbe91","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-20T17:26:32Z","title_canon_sha256":"5022d687ef4d596d506f79047b49527ed014708909a3507703e9bffbbe2e15c9"},"schema_version":"1.0","source":{"id":"2505.14640","kind":"arxiv","version":1}},"canonical_sha256":"816166f3936cba2a25e7293cf018d66b041c13d14c6edcd30bb7686fb18cfb0c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"816166f3936cba2a25e7293cf018d66b041c13d14c6edcd30bb7686fb18cfb0c","first_computed_at":"2026-07-05T11:06:16.894054Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:06:16.894054Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qz+dF7lL6995F2SANkn3jb4i4XA3nVcCaH4QOAU8Poq1AJFx4yFxbnCcrxl5D/shG2NaPrb+FPcFp5g81NLXBg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:06:16.894577Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.14640","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8f5cf868d932ad3de1bbde2853f4a747cef7d910bc5008b225a254c236439a15","sha256:07067a8633c21b301a7e773f104b6f86a34348b4df94fbf1e3c8e54e165f1234"],"state_sha256":"34fe2de60a262c58660cf3509aabd87a34f6134c449ba01370c71350d1138c3d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"m+D0AM9/g/5siQBDom05qb4iGyIIFA5TTbEmYYitddY4imSxyqbtKrw20pmMrBaAGpc52Sz/b0iowJPPIPPNBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T19:21:17.253813Z","bundle_sha256":"54e5afa7a0f6cd8e007e4cf1855a54701d46af081d7b00e082bcf114be8672a8"}}