이진욱님의 블로그

InternLM-XComposer2.5-Reward: A Simple Yet Effective Multi-Modal Reward Model 요약

3IXC2.5-RewardData PreparationReward models are trained using pairwise preference annotations (e.g., prompts x with chosen responses yc and rejected responses yr) that reflect human preferences. While existing public preference data is primarily textual, with limited image and scarce video examples, we train IXC-2.5-Reward using both open-source data and a newly collected dataset to ensure broad..

MASTER: A Multi-Agent System with LLM Specialized MCTS

3.1 PreliminariesBefore introducing our framework, we first present MCTS to clarify the motivation behind our work.  MCTS Coulom (2006) is a widely used planning algorithm and famously employed in AlphaGo Silver et al. (2016). Taking the Game of Go as an example, the algorithm assists in selecting the best possible action in the current state of the board based on their average rewards.  These r..