Tag: agent benchmarking
Browse our exclusive articles!
Efficient LAM Evaluation Aligned with Human Preferences
Discover how the HUMANS benchmark improves large audio model evaluation by aligning with human preferences, reducing costs, and boosting accuracy.
Benchmarking LLM Utility Recovery with User Intent Clarification
Explore how large language models recover utility by clarifying user intent in multi-turn conversations while maintaining safety.
ATBench-Claw & Codex: Benchmarks for Agent Safety
Explore ATBench-Claw and ATBench-Codex, key benchmarks for evaluating and diagnosing agent trajectory safety in OpenClaw and Codex environments.
EnterpriseDocBench: Unified Benchmark for Document AI Pipelines
Discover EnterpriseDocBench, a framework for benchmarking complex multimodal document processing pipelines in enterprise AI with unified evaluation metrics...
Mind-ParaWorld: Evaluating Search Agents in Parallel Worlds
Discover Mind-ParaWorld, a novel framework tackling challenges in evaluating search agents within dynamic, parallel world scenarios for AI models.