RLHF: A comprehensive Survey for Cultural, Multimodal and Low Latency Alignment Methods
This is an incremental survey paper that organizes existing research for AI researchers working on alignment.
This survey synthesizes recent progress in Reinforcement Learning from Human Feedback (RLHF) beyond text-based methods, addressing multi-modal alignment, cultural fairness, and low-latency optimization to provide a roadmap for building more robust, efficient, and equitable AI systems.
Reinforcement Learning from Human Feedback (RLHF) is the standard for aligning Large Language Models (LLMs), yet recent progress has moved beyond canonical text-based methods. This survey synthesizes the new frontier of alignment research by addressing critical gaps in multi-modal alignment, cultural fairness, and low-latency optimization. To systematically explore these domains, we first review foundational algo- rithms, including PPO, DPO, and GRPO, before presenting a detailed analysis of the latest innovations. By providing a comparative synthesis of these techniques and outlining open challenges, this work serves as an essential roadmap for researchers building more robust, efficient, and equitable AI systems.