GRASPrune: Efficient Budgeted Pruning for Large Language Models
Discover GRASPrune, a structured pruning method that reduces large language model size by 50% while maintaining performance and lowering operational costs.
Energy Efficiency in AI-Powered 6G IoT: Centralized vs Decentralized
Explore energy consumption in AI-driven 6G IoT networks, comparing centralized and decentralized learning for optimal sustainability and accuracy.
Evaluating LLM Agents in Capture The Flag Cybersecurity Tests
Discover DeepRed, a benchmark assessing LLM agents' performance in realistic Capture The Flag cybersecurity challenges with partial-credit scoring.
How Large Language Models Show Normative Conformity
Explore how large language models exhibit normative conformity, impacting AI decision-making and group dynamics in multi-agent systems.
Explicit Trait Inference Boosts Multi-Agent Coordination
Discover how Explicit Trait Inference improves multi-agent coordination by tracking partner traits like trust and skill for better decision-making.