Benchmark

RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents featured image

RecreationWorld: Scalable and Verifiable Environments for Hybrid Computer-Use Agents

A five-platform framework and 250-task benchmark for evaluating hybrid computer-use agents that explore, build, and verify software recreations.

recreationworld-team-alibaba-token-hub-alibaba-group
•
MME-Reasoning: A Comprehensive Benchmark for Logical Reasoning in MLLMs featured image

MME-Reasoning: A Comprehensive Benchmark for Logical Reasoning in MLLMs

A comprehensive benchmark specifically designed to evaluate the reasoning capability of MLLMs.

jiakang-yuan
•