AI Paper Atlas

Research map / Video / Video-language understanding

V2P-Bench: Evaluating Video-Language Understanding with Visual Prompts for Better Human-Model Interaction

Yiming Zhao, Yu Zeng, Yukun Qi, YaoYang Liu, Xikun Bao, Lin Chen, Zehui Chen, Qing Miao, Chenxi Liu, Jie Zhao, Feng Zhao

ICLR 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding

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
convert visual prompts to text prompts (ablation)Gemini-2.5-Pro · V2P-Benchaccuracy (%): 69.8 -> 54.7Hurt
convert visual prompts to text prompts (ablation)GPT-4o · V2P-Benchaccuracy (%): 65.4 -> 53.0Hurt
use visual prompts instead of text prompts for users (ablation)user experience study · user study (20 volunteers)completion time (s): 25.2 -> 18.1Improved
use visual prompts instead of text prompts for users (ablation)user experience study · user study (20 volunteers)user satisfaction: 5.3 -> 7.5Improved
fine-tune LLaVA-NeXT on INST-IT visual prompt dataLLaVA-NeXT-7B · V2P-Benchaverage accuracy (%): 46.0 -> 46.3No clear effect
scale model from 7B to 72B (ablation)Qwen2.5-VL · V2P-Benchaverage accuracy (%): 48.1 -> 59.8Improved
use doodle-style visual prompts instead of standard shapes (ablation)Qwen2.5-VL-7B and MiMo-VL-7B · V2P-Bench 217-instance subsetaccuracy (%): decrease of 0.7% and 0.8%Hurt
use arrow instead of rectangle visual prompts (ablation)Qwen2.5-VL-7B · V2P-Bench 217-instance subsetaccuracy (%): 47.3 -> 43.6Hurt
increase video length from short/medium to long (ablation)o1 · V2P-Benchaccuracy (%): 83.9 -> 60.4Hurt
reduce sampled frames from 128 to 4 (ablation)Qwen2.5-VL-7B · V2P-BenchHack Ratio (%): 8.0 -> 18.7Hurt
increase sampled frames from 1 to 64 (ablation)Qwen2.5-VL-7B · V2P-Bench short videos, Temporal Understandingaccuracy (%): 28.6 -> 40.1Improved
increase sampled frames from 1 to 64 (ablation)Qwen2.5-VL-7B · V2P-Bench long videos, Basic Perceptionaccuracy (%): 47.6 -> 43.1Hurt

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