arXiv 2610.31494 · cs.CV · 2026-10-22
AgentVQA: A Unified Benchmark for Agentic Visual Understanding
Advait Gupta, Sachin Konan, Tom Yin, Gabriel Sarch, Zhuang Liu
zlab-princeton
In latest · #307 Added on 2026-10-26
One-line takeaway
AgentVQA is an offline benchmark of 1,000 hand-picked real agent examples from 23 datasets across web agents, robotics, egocentric video, games and spatial planning, for evaluating VLMs in agentic settings.
AgentVQA 是离线评测基准，从 23 个数据集人工挑出 1000 个真实智能体示例，覆盖网页智能体、机器人、第一人称视频、游戏和空间规划五类场景，用来评估视觉语言模型的智能体能力。
Read on arXiv PDF Hugging Face discussion Code Project page
Topics & disciplines
More on Benchmarks & data
- RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversations ▲287
- PremiseCheck: Efficient Fact-Checking of Human and AI Peer Reviews ▲210
- DecisionPivot: A Diagnostic Benchmark for Decision Models in LLM Systems ▲74
- GameHorizon Suite: Multi-Horizon Data and Evaluation in Gameplay ▲73
- More Choices, Fewer Decisions: Ordinal-Scale Bias in JEV-like Direct-Decision Models ▲65
More on Multimodal
- StreamEval: Evaluating Interactive Streaming Video Generation Models 2026-10-20 · ▲649
- MJev-Spatial: System One Spatial Intelligence via Native Attention Matching 2026-10-15 · ▲556
- Omni-IO Skills: Harnessing Your Agent Omni-Native 2026-09-25 · ▲485
- LoopVL: Recurrent Visual Intelligence 2026-09-29 · ▲417
- VideoJev: System One Model for Video Understanding and Interaction 2026-10-13 · ▲375
About this data
Metadata from the arXiv API (verified); upvotes from Hugging Face Daily Papers; citations from Semantic Scholar. Last refreshed 2026-10-26.
The takeaway is written by AIMap from the abstract; the abstract itself is not reproduced — read the original.