VideoHEDGE: Entropy-Based Hallucination Detection for Video-VLMs via Semantic Clustering and Spatiotemporal Perturbations

• URL: http://arxiv.org/abs/2601.08557v1
• Date: Tue, 13 Jan 2026 13:42:05 GMT
• Title: VideoHEDGE: Entropy-Based Hallucination Detection for Video-VLMs via Semantic Clustering and Spatiotemporal Perturbations
• Authors: Sushant Gautam, Cise Midoglu, Vajira Thambawita, Michael A. Riegler, Pål Halvorsen,
• Abstract summary: Hallucinations in video-capable vision models (VideoVLMs) remain frequent and high-confidence.<n>We introduce VideoHEDGE, a modular framework for hallucination detection in question answering.
• Score: 4.509454543418357
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Hallucinations in video-capable vision-language models (Video-VLMs) remain frequent and high-confidence, while existing uncertainty metrics often fail to align with correctness. We introduce VideoHEDGE, a modular framework for hallucination detection in video question answering that extends entropy-based reliability estimation from images to temporally structured inputs. Given a video-question pair, VideoHEDGE draws a baseline answer and multiple high-temperature generations from both clean clips and photometrically and spatiotemporally perturbed variants, then clusters the resulting textual outputs into semantic hypotheses using either Natural Language Inference (NLI)-based or embedding-based methods. Cluster-level probability masses yield three reliability scores: Semantic Entropy (SE), RadFlag, and Vision-Amplified Semantic Entropy (VASE). We evaluate VideoHEDGE on the SoccerChat benchmark using an LLM-as-a-judge to obtain binary hallucination labels. Across three 7B Video-VLMs (Qwen2-VL, Qwen2.5-VL, and a SoccerChat-finetuned model), VASE consistently achieves the highest ROC-AUC, especially at larger distortion budgets, while SE and RadFlag often operate near chance. We further show that embedding-based clustering matches NLI-based clustering in detection performance at substantially lower computational cost, and that domain fine-tuning reduces hallucination frequency but yields only modest improvements in calibration. The hedge-bench PyPI library enables reproducible and extensible benchmarking, with full code and experimental resources available at https://github.com/Simula/HEDGE#videohedge .

Related papers

• CounterVid: Counterfactual Video Generation for Mitigating Action and Temporal Hallucinations in Video-Language Models [66.56549019393042]
  Video-language models (VLMs) achieve strong multimodal understanding but remain prone to hallucinations, especially when reasoning about actions and temporal order.<n>We propose a scalable framework for counterfactual video generation that synthesizes videos differing only in actions or temporal structure while preserving scene context.
  arXiv  Detail & Related papers  (2026-01-08T10:03:07Z)
• TV-RAG: A Temporal-aware and Semantic Entropy-Weighted Framework for Long Video Retrieval and Understanding [14.570869250170139]
  TV-RAG is a training-free architecture that couples temporal alignment with entropy-guided semantics to improve long-video reasoning.<n>By weaving these temporal and semantic signals together, TV-RAG realises a dual-level reasoning routine that can be grafted onto any LVLM without re-training or fine-tuning.
  arXiv  Detail & Related papers  (2025-12-29T14:10:22Z)
• HEDGE: Hallucination Estimation via Dense Geometric Entropy for VQA with Vision-Language Models [4.099133096025821]
  Vision-language models (VLMs) enable open-ended visual question answering but remain prone to hallucinations.<n>We present HEDGE, a unified framework for hallucination detection that combines controlled visual perturbations, semantic clustering, and robust uncertainty metrics.
  arXiv  Detail & Related papers  (2025-11-16T17:16:31Z)
• DUAL-VAD: Dual Benchmarks and Anomaly-Focused Sampling for Video Anomaly Detection [8.294763803639391]
  Video Anomaly Detection (VAD) is critical for surveillance and public safety.<n>Existing benchmarks are limited to either frame-level or video-level tasks.<n>This work introduces a softmax-based frame allocation strategy that prioritizes anomaly-dense segments while maintaining full-video coverage.
  arXiv  Detail & Related papers  (2025-09-15T05:48:22Z)
• VideoMolmo: Spatio-Temporal Grounding Meets Pointing [66.19964563104385]
  VideoMolmo is a model tailored for fine-grained pointing of video sequences.<n>A novel temporal mask fusion employs SAM2 for bidirectional point propagation.<n>To evaluate the generalization of VideoMolmo, we introduce VPoMolS-temporal, a challenging out-of-distribution benchmark spanning five real-world scenarios.
  arXiv  Detail & Related papers  (2025-06-05T17:59:29Z)
• Video-Level Language-Driven Video-Based Visible-Infrared Person Re-Identification [47.40091830500585]
  Video-based Visible-basedInfrared Person Re-Identification (VVIReID) aims to match pedestrian sequences across modalities by extracting modality-in sequence-level features.<n>A framework, video-level language-driven VVI-ReID (VLD), consists of two core modules: inmodality language (IMLP) and spatialtemporal aggregation.
  arXiv  Detail & Related papers  (2025-06-03T04:49:08Z)
• Weakly Supervised Video Anomaly Detection and Localization with Spatio-Temporal Prompts [57.01985221057047]
  This paper introduces a novel method that learnstemporal prompt embeddings for weakly supervised video anomaly detection and localization (WSVADL) based on pre-trained vision-language models (VLMs) Our method achieves state-of-theart performance on three public benchmarks for the WSVADL task.
  arXiv  Detail & Related papers  (2024-08-12T03:31:29Z)
• SIGMA:Sinkhorn-Guided Masked Video Modeling [69.31715194419091]
  Sinkhorn-guided Masked Video Modelling ( SIGMA) is a novel video pretraining method. We distribute features of space-time tubes evenly across a limited number of learnable clusters. Experimental results on ten datasets validate the effectiveness of SIGMA in learning more performant, temporally-aware, and robust video representations.
  arXiv  Detail & Related papers  (2024-07-22T08:04:09Z)
• VaQuitA: Enhancing Alignment in LLM-Assisted Video Understanding [63.075626670943116]
  We introduce a cutting-edge framework, VaQuitA, designed to refine the synergy between video and textual information. At the data level, instead of sampling frames uniformly, we implement a sampling method guided by CLIP-score rankings. At the feature level, we integrate a trainable Video Perceiver alongside a Visual-Query Transformer.
  arXiv  Detail & Related papers  (2023-12-04T19:48:02Z)
• Transform-Equivariant Consistency Learning for Temporal Sentence Grounding [66.10949751429781]
  We introduce a novel Equivariant Consistency Regulation Learning framework to learn more discriminative representations for each video. Our motivation comes from that the temporal boundary of the query-guided activity should be consistently predicted. In particular, we devise a self-supervised consistency loss module to enhance the completeness and smoothness of the augmented video.
  arXiv  Detail & Related papers  (2023-05-06T19:29:28Z)

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.