Prospect Theory Limits in LLMs Under Epistemic Uncertainty
Explore the instability of decision-making in LLMs using Prospect Theory amid epistemic uncertainty and linguistic ambiguity.
ChipSeek: Reinforcement Learning for Optimized Verilog RTL
Discover ChipSeek, a reinforcement learning framework integrating EDA tools to optimize Verilog RTL code for power, performance, and area metrics.
Bayesian Social Deduction Using Graph-Based Language Models
Discover how Bayesian social deduction with graph-informed language models improves AI social reasoning and outperforms humans in complex games.
ROE Framework: AI Learning from Expert & Self Play in StarCraft II
Discover how the ROE framework boosts AI learning in StarCraft II by combining expert insights and self-reflection for superior gameplay strategies.
How Large Language Models Generate Harmful Content
Discover how large language models produce harmful content via a unified mechanism and explore new strategies to improve AI safety and alignment.