AI Paper Atlas

Research map / Video / Video-language understanding

VirtueBench: Evaluating Trustworthiness under Uncertainty in Long Video Understanding

Xueqing Yu, Bohan Li, Yan Li, Zhenheng Yang

CVPR 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 1 citations (OpenAlex)

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 changedWhereMeasuredEffect
remove explicit honesty instruction from prompt (ablation)Qwen3-VL-8B-Instruct · VirtueBench (64 frames)refusal accuracy (%): 36.3 -> 22.8Hurt
remove explicit honesty instruction from prompt (ablation)Qwen3-VL-32B-Instruct · VirtueBench (64 frames)refusal accuracy (%): 60.2 -> 33.7Hurt
remove explicit honesty instruction from prompt (ablation)Qwen3-VL-32B-Thinking · VirtueBench (64 frames)refusal accuracy (%): 43.1 -> 9.4Hurt
increase input frames from 64 to 1024Qwen3VL-32B · VirtueBenchoverall accuracy (%): 50.83 -> 32.54Hurt
increase input frames from 64 to 1024Gemini-2.5-Flash · VirtueBenchoverall accuracy (%): 58.96 -> 53.66Hurt
RL/reasoning training (Mimo-VL-RL vs base)Mimo-VL-7B · VirtueBench (64 frames)refusal accuracy (%): 32.35 -> 40.78Improved
reasoning-optimized GPT-5 vs GPT-5-chatGPT-5 · VirtueBench (64 frames)refusal accuracy (%): 19.80 -> 33.53Improved
larger model size Qwen2.5-VL 7B to 72BQwen2.5-VL · VirtueBench (64 frames)refusal accuracy (%): 53.33 -> 80.20Improved
thinking variant of Qwen3-VL-32B vs instructQwen3-VL-32B · VirtueBench (64 frames)refusal accuracy (%): 60.20 -> 43.14Hurt
RL training on InternVL3.5-38BInternVL3.5-38B · VirtueBench (64 frames)refusal accuracy (%): 0.59 -> 0.59No clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy