AgentRewardBench

What is AgentRewardBench?

AgentRewardBench, the first benchmark to assess the effectiveness of LLM judges for evaluating web agents. AgentRewardBench contains 1302 trajectories across 5…

Released

2025-04-11

Evaluates

推理, Reasoning, 智能体, Agent, 任务执行, Task Execution, 不支持, Unsupported

Openness

restricted

Importer

OpenCompass Hub

Reported scores

0

Source provenance

• Original evidence https://hub.opencompass.org.cn/dataset-detail/AgentRewardBench

AgentRewardBench paper and dataset

• Paper https://arxiv.org/abs/2504.08942
• Dataset https://huggingface.co/datasets/McGill-NLP/agent-reward-bench

Which sources cite AgentRewardBench?

• AgentRewardBench

Related 推理, Reasoning, 智能体, Agent, 任务执行, Task Execution, 不支持, Unsupported benchmarks

• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• ActiveFly-Bench
• Ada-LEval
• AdvancedMathBench
• AECBench