Tag: tool use
Browse our exclusive articles!
Optimizing Small Language Models with Agent Paradigms
Explore how agent paradigms enhance small language models, overcoming limits in knowledge and reasoning for scalable, efficient NLP deployment.
Enhancing LLM Tool Use via Automated Feedback Environments
Discover how automated environments and feedback-driven training improve large language models' tool use and task execution precision.
UniToolCall: Standardizing Tool-Use for LLM Agents
UniToolCall unifies tool-use representation, data, and evaluation to boost LLM agents' performance with over 22K tools and diverse interaction patterns.
The Amazing Agent Race: Navigating Complex Tool Use
Explore The Amazing Agent Race benchmark testing LLM agents' navigation and multi-step tool use with DAG puzzles and real-time validation.
WildToolBench: Real-World Benchmark for LLM Tool Use
Discover WildToolBench, a new benchmark revealing the real-world challenges LLMs face in tool use with complex user interactions and low accuracy rates.