IEEE

Reinforcement Learning-powered Effectiveness and Efficiency Few-shot Jailbreaking Attack LLMs

Abstract: The widespread use of large language models (LLMs) has brought about security risks, including biases, discrimination, and ethical concerns. Reinforcement Learning from Human Feedback (RLHF) ...
IEEE

DemoCraft: Using In-Context Learning to Improve Code Generation in Large Language Models

Abstract: Producing executable code from natural-language directives via Large Language Models (LLMs) involves obstacles like semantic uncertainty and the requirement for task-focused context ...

From Research To Reality: How RL Environments Can Unlock The Next Wave Of AI Agents

A reinforcement learning environment is a fail-safe digital practice room where an agent can afford to make mistakes and ...
GitHub

DARE: dLLM Alignment and Reinforcement Executor

Easy extension of diverse RL algorithms for dLLMs Easy extension of extra benchmark evaluations for dLLMs Easy integration of popular and upcoming dLLM infras and HuggingFace weights DARE is a work in ...