Paper Citation Record · LEDGER

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA

As of 12 August 2026, this Paper Citation Record lists 30 of 30 outbound references and 2 inbound Pith citation observations for arXiv:2606.30220.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2606.30220 v1

Coverage vector

measured 30 of 30 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-06-30T06:14:11.109840Z

measured 32 of 32 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-11T06:34:44.6726+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-01T10:15:13.957224Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: cited_works

Reference resolution

30 of 30 outbound references displayed

verified exact17
verified fuzzy0
unresolved9
parse uncertain0
malformed identifier1
metadata mismatch3

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 1e364f0f-4414-484a-8593-95681bd6caf2 · outbound

This paper cites Flamingo: a Visual Language Model for Few-Shot Learning.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Flamingo: a Visual Language Model for Few-Shot Learning

Reference 1

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.544223Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation daa1d942-08ea-4bb0-810c-d6dd88364ec3 · outbound

This paper cites Aofei Chang, Le Huang, Parminder Bhatia, Taha Kass-Hout, Fenglong Ma, and Cao Xiao.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Aofei Chang, Le Huang, Parminder Bhatia, Taha Kass-Hout, Fenglong Ma, and Cao Xiao

Reference 2

Resolution: metadata mismatch; arxiv_id, observed 2026-06-30T06:14:18.547646Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 5fc94880-bb2f-4090-b09a-692695eff023 · outbound

This paper cites Qwen2.5-VL Technical Report.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Qwen2.5-VL Technical Report

Reference 3

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.554272Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 1d7aac15-7aa4-4732-8fc6-f2d2ddde606e · outbound

This paper cites train on the test set.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA train on the test set

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-06-30T06:14:18.557284Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 83c8b099-376e-4940-80a7-e7f2feba3c14 · outbound

This paper cites Are We on the Right Way for Evaluating Large Vision-Language Models?.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Are We on the Right Way for Evaluating Large Vision-Language Models?

Reference 5

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.550293Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation c3ab84cc-4b58-4d19-a6bd-b6b31f6be870 · outbound

This paper cites Words or vision: Do vision-language models have blind faith in text? In 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Words or vision: Do vision-language models have blind faith in text? In 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp

Reference 6

Resolution: unresolved; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b9e2e139-f95f-4e12-9a41-46bbe689c212 · outbound

This paper cites Freeman, Frédo Durand, Eli Shechtman, and Xun Huang.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Freeman, Frédo Durand, Eli Shechtman, and Xun Huang

Reference 7

Resolution: metadata mismatch; arxiv_id, observed 2026-06-30T06:14:17.812087Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 8eb521de-6f96-46a2-b234-cb41c556b7b4 · outbound

This paper cites Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Video-MME: The First-Ever Comprehensive Evaluation Benchmark of Multi-modal LLMs in Video Analysis

Reference 8

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.598593Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation ed81f1fa-7da7-435f-9647-6c9504e00fd6 · outbound

This paper cites Do Vision Language Models Need to Process Image Tokens?.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Do Vision Language Models Need to Process Image Tokens?

Reference 9

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.605692Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 68879b39-a139-4fad-a962-e2e03b9870cd · outbound

This paper cites Accidentbench: Benchmarking multimodal understanding and reasoning in vehicle accidents and beyond.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Accidentbench: Benchmarking multimodal understanding and reasoning in vehicle accidents and beyond

Reference 10

Resolution: verified exact; arxiv_id, observed 2026-06-30T06:14:18.599307Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation a0715a46-5816-47e9-9d8a-debf9fbb15c5 · outbound

This paper cites VRU-Accident: A Vision-Language Benchmark for Video Question Answering and Dense Captioning for Accident Scene Understanding.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA VRU-Accident: A Vision-Language Benchmark for Video Question Answering and Dense Captioning for Accident Scene Understanding

Reference 11

Resolution: verified exact; arxiv_id, observed 2026-06-30T06:14:18.592347Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 468a8dbc-ae18-4519-a684-ffc81b210f9d · outbound

This paper cites an unresolved cited work.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Unresolved cited work

Reference 12

Resolution: verified exact; arxiv_id, observed 2026-06-30T06:14:18.602790Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 81997d68-dc64-4062-a352-15a48ce81e85 · outbound

This paper cites LLaVA-OneVision: Easy Visual Task Transfer.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA LLaVA-OneVision: Easy Visual Task Transfer

Reference 13

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.578203Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation bae40ca8-26ad-4ca5-92d8-138f00f4da29 · outbound

This paper cites Understanding language prior of lvlms by contrasting chain-of-embedding.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Understanding language prior of lvlms by contrasting chain-of-embedding

Reference 14

Resolution: verified exact; arxiv_id, observed 2026-06-30T06:14:18.587644Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 365b3d41-34c0-4c51-bf09-e04b99c0b6c3 · outbound

This paper cites EgoSchema: A Diagnostic Benchmark for Very Long-form Video Language Understanding.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA EgoSchema: A Diagnostic Benchmark for Very Long-form Video Language Understanding

Reference 15

Resolution: verified exact; arxiv_id, observed 2026-06-30T06:14:18.581710Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 6d595c44-01a6-4ad5-bd4e-a7b700ece113 · outbound

This paper cites Arbitration Failure, Not Perceptual Blindness: How Vision-Language Models Resolve Visual-Linguistic Conflicts.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Arbitration Failure, Not Perceptual Blindness: How Vision-Language Models Resolve Visual-Linguistic Conflicts

Reference 16

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.572004Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation e61ca1cd-6bcb-4d80-80f7-7efbe79f15d5 · outbound

This paper cites LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA LongVideoBench: A Benchmark for Long-context Interleaved Video-Language Understanding

Reference 17

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.602974Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 28ac769d-8bc6-4dff-b4c4-b4d879a81dd3 · outbound

This paper cites Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data, and Metric Perspectives.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Are VLMs Ready for Autonomous Driving? An Empirical Study from the Reliability, Data, and Metric Perspectives

Reference 18

Resolution: metadata mismatch; arxiv_id, observed 2026-06-30T06:14:18.584457Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation ef53ed70-ff35-49b9-ac16-710152423cf0 · outbound

This paper cites Beyond accuracy: Evaluating visual grounding in multimodal medical reasoning.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Beyond accuracy: Evaluating visual grounding in multimodal medical reasoning

Reference 19

Resolution: verified exact; arxiv_id, observed 2026-06-30T06:14:18.585953Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation bf63b4d9-f67e-4861-b864-62478fbce937 · outbound

This paper cites Watch Before You Answer: Learning from Visually Grounded Post-Training.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Watch Before You Answer: Learning from Visually Grounded Post-Training

Reference 20

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.595304Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation b1a90ab7-aeb1-4544-b5e2-f6ecaa9d7cee · outbound

This paper cites MLVU: Benchmarking Multi-task Long Video Understanding.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA MLVU: Benchmarking Multi-task Long Video Understanding

Reference 21

Resolution: verified exact; local_arxiv, observed 2026-06-30T06:14:18.571984Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 242a7a44-f182-4792-8574-ec9c43acc111 · outbound

This paper cites Dataset Videos Avg.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Dataset Videos Avg

Reference 22

Resolution: unresolved; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0b2e5574-9611-4483-aefc-3b6e98ab426b · outbound

This paper cites an unresolved cited work.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Unresolved cited work

Reference 23

Resolution: unresolved; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d2800445-960d-429d-99a4-d755904df3dc · outbound

This paper cites an unresolved cited work.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Unresolved cited work

Reference 24

Resolution: unresolved; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7d99861-7fcb-4e90-a1e9-4d1d78276ab7 · outbound

This paper cites an unresolved cited work.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Unresolved cited work

Reference 25

Resolution: unresolved; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f9db1599-f84a-4884-942b-65b412814a24 · outbound

This paper cites In multimodal systems this manifests asmodality collapse, where language dominates prediction regardless of visual input (Sim et al., 2025; Deng et al., 2025).

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA In multimodal systems this manifests asmodality collapse, where language dominates prediction regardless of visual input (Sim et al., 2025; Deng et al., 2025)

Reference 26

Resolution: unresolved; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d0f20afc-500f-476b-88b3-3e6da75e40cc · outbound

This paper cites Existing auditing approaches either apply binary filtering by removing questions a blind model answers correctly (Asadi et al., 2026; Zhang et al.,.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Existing auditing approaches either apply binary filtering by removing questions a blind model answers correctly (Asadi et al., 2026; Zhang et al.,

Reference 27

Resolution: unresolved; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f94cb3ce-3bcb-4be5-a8b0-c016635fa455 · outbound

This paper cites Binary filtering discards visually ambiguous but grounded questions; dataset-level metrics cannot identify which individual questions are shortcut-prone.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Binary filtering discards visually ambiguous but grounded questions; dataset-level metrics cannot identify which individual questions are shortcut-prone

Reference 28

Resolution: unresolved; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cc854ff6-84d6-4973-b38b-1027dca88007 · outbound

This paper cites Videos can reach approximately 16 minutes, posing a practical challenge for models under a fixed frame budget.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA Videos can reach approximately 16 minutes, posing a practical challenge for models under a fixed frame budget

Reference 29

Resolution: malformed identifier; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 07379e89-a66b-4d84-8454-3d7042e13d9b · outbound

This paper cites A substantial share of what these benchmarks measure is answerable without any video.

From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA A substantial share of what these benchmarks measure is answerable without any video

Reference 30

Resolution: unresolved; no resolver link, observed 2026-06-30T06:14:11.109840Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation fb9a72ae-3a9b-467b-b250-5d8640e3197a · inbound

Visual Credit Audit for Multimodal Spatial Reasoning cites this paper.

Visual Credit Audit for Multimodal Spatial Reasoning From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA

Reference 7

Resolution: unresolved; no resolver link, observed 2026-07-30T12:08:02.361140Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 08613a70-6534-43ac-805b-5d14fc85d25f · inbound

Visual Credit Audit for Multimodal Spatial Reasoning cites this paper.

Visual Credit Audit for Multimodal Spatial Reasoning From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-01T10:15:13.957224Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.