VADER: Adaptive Debiasing for Hallucination Mitigation in Video Large Language Models

• URL: http://arxiv.org/abs/2608.08622v1
• Date: Sun, 09 Aug 2026 10:09:11 GMT
• Title: VADER: Adaptive Debiasing for Hallucination Mitigation in Video Large Language Models
• Abstract summary: Video-Adaptive Debiasing via Evidence Reweighting (VADER) is a training-free framework with two complementary modules.<n>Visual Focus Reallocation (VFR) automatically instantiates an intervention policy for each video-question input.<n> Selective Evidence Erasure (SEE) independently masks high-importance visual tokens in every frame.
• Score: 30.792102059830274
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large vision-language models (LVLMs) have demonstrated strong performance in open-ended video understanding, yet they remain prone to fluent responses unsupported by video evidence. Existing training-free methods typically apply a globally fixed visual intervention or construct a contrastive branch through input perturbation. The former cannot accommodate video-dependent fusion paths, while the latter can be compensated by cross-frame redundancy. We therefore propose Video-Adaptive Debiasing via Evidence Reweighting (VADER), a training-free framework with two complementary modules. Visual Focus Reallocation (VFR) automatically instantiates an intervention policy for each video-question input: it diagnoses layer-wise visual-to-text evidence flow, determines where to intervene, and derives how strongly to reallocate pre-softmax attention from system-token to video-token blocks. Selective Evidence Erasure (SEE) independently masks high-importance visual tokens in every frame, constructing a prior-biased branch that is difficult to compensate through neighboring frames. Contrastive decoding then down-weights predictions that remain confident after selective evidence erasure. Across multiple VideoLLMs, VADER yields substantial improvements on event-level grounding and temporal consistency; on LLaVA-Video-7B, it reaches 72.60% accuracy on EventHallusion.

Related papers

• VideoScout: Learning Agentic Active Exploration with Adaptive Reasoning Pacing for Long Video Understanding [61.00809248898572]
  Long video understanding is a problem in which an agent reads the video turn by turn along the temporal axis, deciding at each turn how fast to watch, what evidence to retain, when to revisit uncertain segments, and when to stop and answer.<n>We propose textbfVideoScout, a multi-turn reasoning agent that instantiates the SEA paradigm through adaptive reasoning pacing.<n>Experiments on long video understanding and reasoning benchmarks demonstrate that our 7B model achieves strong performance compared with existing trained 7B agentic models.
  arXiv  Detail & Related papers  (2026-09-14T14:04:03Z)
• From Accuracy to Visual Dependence: Auditing and Filtering Modality Collapse in Traffic VideoQA [44.46419663487345]
  High benchmark accuracy does not guarantee genuine use of visual evidence.<n>We study this problem in traffic accident Video Question Answering (VideoQA)<n>We propose an instance-level Shortcut Score that combines text-only confidence with visual necessity signals.
  arXiv  Detail & Related papers  (2026-06-29T12:33:15Z)
• Reflect-R1: Evidence-Driven Reflection for Self-Correction in Long Video Understanding [37.607009528052565]
  This work proposes Reflect-R1, the first Evidence-Driven self-correction framework for long video understanding.<n>The framework constructs a three-stage pipeline consisting of intuition, verification, and arbitration.<n>Our method significantly improves the genuine rectification rate and enables authentic self-correction strictly grounded in objective evidence.
  arXiv  Detail & Related papers  (2026-06-26T10:15:36Z)
• Reasoning as Intersection: Consensus-Frame Alignment for Visual Focus in Video-MLLMs [81.04673240949074]
  Reinforcement learning has improved the reasoning ability of large language models.<n>Applying outcome-only rewards to video multimodal large language models provides limited guidance on which visual evidence should support the answer.<n>We introduce Consensus Frame GRPO, a temporal-annotation-free process-level reward framework for evidence-aware video reasoning.
  arXiv  Detail & Related papers  (2026-06-16T19:42:54Z)
• STORM: Internalized Modeling for Spatial-Temporal Reasoning in Video-Language Models [81.32710031596591]
  Video reasoning tasks require tracking motion, temporal order, and evolving visual states across frames.<n>Existing methods built on vision-language models (LVLMs) often address this challenge by externalizing reasoning through chain-of-thought (CoT)<n>We propose STORMS, a two-stage framework that teaches LVLMs to reason through bounded continuous latent trajectories instead of explicit textual CoT.
  arXiv  Detail & Related papers  (2026-05-25T16:33:00Z)
• Relaxing Anchor-Frame Dominance for Mitigating Hallucinations in Video Large Language Models [22.554751069064906]
  Recent Video Large Language Models (Video-LLMs) have demonstrated strong capability in video understanding, yet they still suffer from hallucinations.<n>We investigate a decoder-side phenomenon in which the model exhibits a temporally imbalanced concentration pattern.<n>We propose Decoder-side Temporal Rebalancing (DTR), a training-free, layer-selective inference method that rebalances temporal evidence allocation in middle-to-late decoder layers.
  arXiv  Detail & Related papers  (2026-04-14T11:05:42Z)
• Diffusion-DRF: Differentiable Reward Flow for Video Diffusion Fine-Tuning [72.16213872139748]
  Diffusion-DRF is a differentiable reward flow for fine-tuning video diffusion models.<n>It backpropagates VLM feedback through the diffusion denoising chain.<n>It improves video quality and semantic alignment while mitigating reward hacking and collapse.
  arXiv  Detail & Related papers  (2026-01-07T18:05:08Z)
• SeViCES: Unifying Semantic-Visual Evidence Consensus for Long Video Understanding [36.30263540665245]
  We propose a framework for effective and reliable long video understanding.<n>SeViCES is training-free and model-agnostic, and introduces two key components.<n> experiments on long video understanding benchmarks show that SeViCES consistently outperforms state-of-the-art methods in both accuracy and robustness.
  arXiv  Detail & Related papers  (2025-10-23T14:55:28Z)
• Unlocking Vision-Language Models for Video Anomaly Detection via Fine-Grained Prompting [17.850029260662648]
  We propose ASK-Hint, a structured prompting framework for video anomaly detection.<n>Our approach organizes prompts into semantically coherent groups and formulates fine-grained guiding questions.<n>Experiments on UCF-Crime and XD-Violence show that ASK-Hint consistently improves AUC over prior baselines.
  arXiv  Detail & Related papers  (2025-10-02T16:06:31Z)
• Video-LLMs with Temporal Visual Screening [59.18455762289321]
  Temporal Visual Screening (TVS) is a new task that universally pre-processes video question answering and instruction tuning data.<n>TVS is formulated as a modular front-end adapter task that can be seamlessly integrated into both Video Instruction Tuning (training) and Video Question Answering (inference) pipelines.<n> Experiments demonstrate that incorporating TVS yields relative gains of 7.33% (training) and 34.6% (inference)
  arXiv  Detail & Related papers  (2025-08-27T14:33:32Z)
• ViaRL: Adaptive Temporal Grounding via Visual Iterated Amplification Reinforcement Learning [68.76048244253582]
  We introduce ViaRL, the first framework to leverage rule-based reinforcement learning (RL) for optimizing frame selection in video understanding.<n>ViaRL utilizes the answer accuracy of a downstream model as a reward signal to train a frame selector through trial-and-error.<n>ViaRL consistently delivers superior temporal grounding performance and robust generalization across diverse video understanding tasks.
  arXiv  Detail & Related papers  (2025-05-21T12:29:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.