Research map / Video / Video generation
No Place to Hide: Benchmarking Video Hallucination with Background-Controlled Pairs
Haojian Huang, Harold Haodong Chen, Meng Luo, Junjia Du, Shanqing Xu, Ziheng Chen, Yanxiang Huang, Yinchuan Li, Yingcong Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use closed-source Gemini-2.5-Pro instead of open-source models | LVMs · VidPair-Halluc binary | wAcc: 49.15 vs best open-source 41.66 | Improved |
| reasoning-tuned ThinkLite-VL instead of Qwen2.5-VL-Instruct (same base) | Qwen2.5-VL 7B · VidPair-Halluc binary | wAcc: 41.66 -> 31.86 | Hurt |
| pair videos with highly similar backgrounds instead of text-only pairs | LVMs · VidPair-Halluc vs VidHalluc | accuracy | Hurt |
| controlled backgrounds (VidPair-Halluc) vs VidHalluc | most LVMs · VidPair-Halluc vs VidHalluc | accuracy | Hurt |
| newer video generator Wan2.2 vs earlier | PairFlow · synthesis human pass rate | pass@1: about 63% -> about 83% | Improved |
Closest papers by meaning
- Learning to Decode Against Compositional Hallucination in Video Multimodal Large Language Models (ICML 2026 · similarity 0.92)
- MHBench: Demystifying Motion Hallucination in VideoLLMs (AAAI 2025 · similarity 0.91)
- VidHalluc: Evaluating Temporal Hallucinations in Multimodal Large Language Models for Video Understanding (CVPR 2025 · similarity 0.91)
- MASH-VLM: Mitigating Action-Scene Hallucination in Video-LLMs through Disentangled Spatial-Temporal Representations (CVPR 2025 · similarity 0.91)
- Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering (NeurIPS 2025 · similarity 0.90)
- Analyzing and Mitigating Object Hallucination: A Training Bias Perspective (AAAI 2026 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
