Tag: agent benchmarking
Browse our exclusive articles!
MAS-Bench: Benchmark for Hybrid Mobile GUI Agents
Discover MAS-Bench, a unified benchmark for evaluating shortcut-augmented hybrid mobile GUI agents with 139 tasks and 9 metrics.
Orak Benchmark: Train & Evaluate LLM Agents in Video Games
Discover Orak, a benchmark for training and evaluating LLM agents across diverse video games with fine-tuning datasets and versatile game coverage.
LongCoT Benchmark: Advancing Long-Horizon Chain-of-Thought AI
Discover LongCoT, a benchmark with 2,500 expert problems to test and improve long-horizon chain-of-thought reasoning in AI models.
LiveClawBench: Benchmarking LLM Agents on Real Tasks
Discover LiveClawBench, a benchmark evaluating LLM agents on complex, real-world assistant tasks using a unique Triple-Axis Complexity Framework.
RiskWebWorld: Benchmarking GUI Agents for E-commerce Risk
Discover RiskWebWorld, a realistic benchmark for testing GUI agents in complex e-commerce risk management tasks across multiple domains.