{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:YNQJL4CMKFECWAPFYQ6KXDPJKD","short_pith_number":"pith:YNQJL4CM","canonical_record":{"source":{"id":"2603.27259","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-28T12:44:19Z","cross_cats_sorted":[],"title_canon_sha256":"3ca3f6151f7d08480faf12c4676262d71303d8b81b2330b5e919c0d9e9c76115","abstract_canon_sha256":"9b4a89f2c7c43f0f960d61d77cc48477fdefe91ef8f2eef88e3cedeea7cb2d5c"},"schema_version":"1.0"},"canonical_sha256":"c36095f04c51482b01e5c43cab8de950e4981d45e47e73177019011181475703","source":{"kind":"arxiv","id":"2603.27259","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.27259","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"arxiv_version","alias_value":"2603.27259v3","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.27259","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"pith_short_12","alias_value":"YNQJL4CMKFEC","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"pith_short_16","alias_value":"YNQJL4CMKFECWAPF","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"pith_short_8","alias_value":"YNQJL4CM","created_at":"2026-06-23T01:12:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:YNQJL4CMKFECWAPFYQ6KXDPJKD","target":"record","payload":{"canonical_record":{"source":{"id":"2603.27259","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-28T12:44:19Z","cross_cats_sorted":[],"title_canon_sha256":"3ca3f6151f7d08480faf12c4676262d71303d8b81b2330b5e919c0d9e9c76115","abstract_canon_sha256":"9b4a89f2c7c43f0f960d61d77cc48477fdefe91ef8f2eef88e3cedeea7cb2d5c"},"schema_version":"1.0"},"canonical_sha256":"c36095f04c51482b01e5c43cab8de950e4981d45e47e73177019011181475703","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T01:12:04.172903Z","signature_b64":"8k4ZkCt1i8+XYIIPBYoH/TNr3+dSLXpcVEyfTWh7IajitPqBDneuBJbaKMLfzR2HnIJ/Km3ViJD2ArX5c4UODQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c36095f04c51482b01e5c43cab8de950e4981d45e47e73177019011181475703","last_reissued_at":"2026-06-23T01:12:04.172364Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T01:12:04.172364Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.27259","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":"2026-06-23T01:12:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VfjH8AMjAcREy5uMb0ozI8KM8kKEo1IxSEsx5tgkwV7fBQmoMrAUSfJLWdl5sdJHcnM6rqyMlURIsSFibGVABA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-25T06:04:09.892863Z"},"content_sha256":"b424360951ab9cf17004340d6634662c14d9694d1913ee8105bb38dce294d574","schema_version":"1.0","event_id":"sha256:b424360951ab9cf17004340d6634662c14d9694d1913ee8105bb38dce294d574"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:YNQJL4CMKFECWAPFYQ6KXDPJKD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Seeing the Scene Matters: Revealing Forgetting in Video Understanding Models with a Scene-Aware Long-Video Benchmark","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"Vision-language models forget long-range scene context in videos, shown by a new benchmark with sharp accuracy drops.","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chao Li, Chenglam Ho, Hao Chen, Jinping Wang, Seng Nam Chen, Xinyu Mao, Yu Zhang","submitted_at":"2026-03-28T12:44:19Z","abstract_excerpt":"Long video understanding (LVU) remains a core challenge in multimodal learning. Although recent vision-language models (VLMs) have made notable progress, existing benchmarks mainly focus on either fine-grained perception or coarse summarization, offering limited insight into temporal understanding over long contexts. In this work, we define a scene as a coherent segment of a video in which both visual and semantic contexts remain consistent, aligning with human perception. This leads us to a key question: can current VLMs reason effectively over long, scene-level contexts? To answer this, we i"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our evaluation reveals a sharp drop in accuracy when VLMs attempt to answer scene-level questions, indicating significant forgetting of long-range context.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the authors' definition of a scene as a coherent segment with consistent visual and semantic contexts accurately isolates long-range forgetting, and that the benchmark questions do not introduce other confounds in video selection or question design.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SceneBench shows VLMs lose accuracy on scene-level questions in long videos due to forgetting, and Scene-RAG retrieval improves performance by 2.5%.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Vision-language models forget long-range scene context in videos, shown by a new benchmark with sharp accuracy drops.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"a917d5f8448758645675e74b0e4a912d18ccb4bfee6a590719933b28662f355d"},"source":{"id":"2603.27259","kind":"arxiv","version":3},"verdict":{"id":"b363ad80-56a4-4b57-bcbe-4a1a8a10ddcb","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T22:02:10.306737Z","strongest_claim":"Our evaluation reveals a sharp drop in accuracy when VLMs attempt to answer scene-level questions, indicating significant forgetting of long-range context.","one_line_summary":"SceneBench shows VLMs lose accuracy on scene-level questions in long videos due to forgetting, and Scene-RAG retrieval improves performance by 2.5%.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the authors' definition of a scene as a coherent segment with consistent visual and semantic contexts accurately isolates long-range forgetting, and that the benchmark questions do not introduce other confounds in video selection or question design.","pith_extraction_headline":"Vision-language models forget long-range scene context in videos, shown by a new benchmark with sharp accuracy drops."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2603.27259/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":"b363ad80-56a4-4b57-bcbe-4a1a8a10ddcb"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-23T01:12:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bFon7043KI7wxc+OQxYlo3hbmCLKuPpm5i0vLTFs/7aMyr/PEG7ddGBh5+U6kjppRMnEcqHU396E3lGAP+sXCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-25T06:04:09.893351Z"},"content_sha256":"50973c25b5a8f64c212a55a23396ece83a56611dd28d7d40a1e012da39114a36","schema_version":"1.0","event_id":"sha256:50973c25b5a8f64c212a55a23396ece83a56611dd28d7d40a1e012da39114a36"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YNQJL4CMKFECWAPFYQ6KXDPJKD/bundle.json","state_url":"https://pith.science/pith/YNQJL4CMKFECWAPFYQ6KXDPJKD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YNQJL4CMKFECWAPFYQ6KXDPJKD/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":"2026-06-25T06:04:09Z","links":{"resolver":"https://pith.science/pith/YNQJL4CMKFECWAPFYQ6KXDPJKD","bundle":"https://pith.science/pith/YNQJL4CMKFECWAPFYQ6KXDPJKD/bundle.json","state":"https://pith.science/pith/YNQJL4CMKFECWAPFYQ6KXDPJKD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YNQJL4CMKFECWAPFYQ6KXDPJKD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YNQJL4CMKFECWAPFYQ6KXDPJKD","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":"9b4a89f2c7c43f0f960d61d77cc48477fdefe91ef8f2eef88e3cedeea7cb2d5c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-28T12:44:19Z","title_canon_sha256":"3ca3f6151f7d08480faf12c4676262d71303d8b81b2330b5e919c0d9e9c76115"},"schema_version":"1.0","source":{"id":"2603.27259","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.27259","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"arxiv_version","alias_value":"2603.27259v3","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.27259","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"pith_short_12","alias_value":"YNQJL4CMKFEC","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"pith_short_16","alias_value":"YNQJL4CMKFECWAPF","created_at":"2026-06-23T01:12:04Z"},{"alias_kind":"pith_short_8","alias_value":"YNQJL4CM","created_at":"2026-06-23T01:12:04Z"}],"graph_snapshots":[{"event_id":"sha256:50973c25b5a8f64c212a55a23396ece83a56611dd28d7d40a1e012da39114a36","target":"graph","created_at":"2026-06-23T01:12:04Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Our evaluation reveals a sharp drop in accuracy when VLMs attempt to answer scene-level questions, indicating significant forgetting of long-range context."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the authors' definition of a scene as a coherent segment with consistent visual and semantic contexts accurately isolates long-range forgetting, and that the benchmark questions do not introduce other confounds in video selection or question design."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SceneBench shows VLMs lose accuracy on scene-level questions in long videos due to forgetting, and Scene-RAG retrieval improves performance by 2.5%."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Vision-language models forget long-range scene context in videos, shown by a new benchmark with sharp accuracy drops."}],"snapshot_sha256":"a917d5f8448758645675e74b0e4a912d18ccb4bfee6a590719933b28662f355d"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2603.27259/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Long video understanding (LVU) remains a core challenge in multimodal learning. Although recent vision-language models (VLMs) have made notable progress, existing benchmarks mainly focus on either fine-grained perception or coarse summarization, offering limited insight into temporal understanding over long contexts. In this work, we define a scene as a coherent segment of a video in which both visual and semantic contexts remain consistent, aligning with human perception. This leads us to a key question: can current VLMs reason effectively over long, scene-level contexts? To answer this, we i","authors_text":"Chao Li, Chenglam Ho, Hao Chen, Jinping Wang, Seng Nam Chen, Xinyu Mao, Yu Zhang","cross_cats":[],"headline":"Vision-language models forget long-range scene context in videos, shown by a new benchmark with sharp accuracy drops.","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-28T12:44:19Z","title":"Seeing the Scene Matters: Revealing Forgetting in Video Understanding Models with a Scene-Aware Long-Video Benchmark"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.27259","kind":"arxiv","version":3},"verdict":{"created_at":"2026-05-14T22:02:10.306737Z","id":"b363ad80-56a4-4b57-bcbe-4a1a8a10ddcb","model_set":{"reader":"grok-4.3"},"one_line_summary":"SceneBench shows VLMs lose accuracy on scene-level questions in long videos due to forgetting, and Scene-RAG retrieval improves performance by 2.5%.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Vision-language models forget long-range scene context in videos, shown by a new benchmark with sharp accuracy drops.","strongest_claim":"Our evaluation reveals a sharp drop in accuracy when VLMs attempt to answer scene-level questions, indicating significant forgetting of long-range context.","weakest_assumption":"That the authors' definition of a scene as a coherent segment with consistent visual and semantic contexts accurately isolates long-range forgetting, and that the benchmark questions do not introduce other confounds in video selection or question design."}},"verdict_id":"b363ad80-56a4-4b57-bcbe-4a1a8a10ddcb"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:b424360951ab9cf17004340d6634662c14d9694d1913ee8105bb38dce294d574","target":"record","created_at":"2026-06-23T01:12:04Z","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":"9b4a89f2c7c43f0f960d61d77cc48477fdefe91ef8f2eef88e3cedeea7cb2d5c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-03-28T12:44:19Z","title_canon_sha256":"3ca3f6151f7d08480faf12c4676262d71303d8b81b2330b5e919c0d9e9c76115"},"schema_version":"1.0","source":{"id":"2603.27259","kind":"arxiv","version":3}},"canonical_sha256":"c36095f04c51482b01e5c43cab8de950e4981d45e47e73177019011181475703","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c36095f04c51482b01e5c43cab8de950e4981d45e47e73177019011181475703","first_computed_at":"2026-06-23T01:12:04.172364Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T01:12:04.172364Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8k4ZkCt1i8+XYIIPBYoH/TNr3+dSLXpcVEyfTWh7IajitPqBDneuBJbaKMLfzR2HnIJ/Km3ViJD2ArX5c4UODQ==","signature_status":"signed_v1","signed_at":"2026-06-23T01:12:04.172903Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.27259","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:b424360951ab9cf17004340d6634662c14d9694d1913ee8105bb38dce294d574","sha256:50973c25b5a8f64c212a55a23396ece83a56611dd28d7d40a1e012da39114a36"],"state_sha256":"1149b8a87aacceea51ca6fd5d1a1e4ae3814c68951a30cb18aa324da4e4ebca6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BPGV26YFBUh9zGEygf28aKVZW1s9co5f63JyAmzU7m0ehDSBdbf0AGnZ330q+87HeZiW0BTKnyPR7C/sBFojDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-25T06:04:09.895625Z","bundle_sha256":"39b27fa5cb3f8ab111d62b526630a699a15af5afa476bb5f082e6a1892b21780"}}