Preference tuning is a crucial process for aligning deep generative models with human
preferences. This survey offers a thorough overview of recent advancements in preference …

Reinforcement learning from human feedback (RLHF) has proven effective in enhancing the
instruction-following capabilities of large language models; however, it remains …

Self-improvement in multimodal large language models (MLLMs) is crucial for enhancing
their reliability and robustness. However, current methods often rely heavily on MLLMs …

Direct Preference Optimization (DPO) is effective for aligning large language models (LLMs),
but when applied to multimodal models (MLLMs), it often favors text over image information …

Multimodal large language models (MLLMs) excel at multimodal perception and
understanding, yet their tendency to generate hallucinated or inaccurate responses …

The rapid advancement of large multimodal models (LMMs) has revolutionized the field of
deep learning, enabling sophisticated understanding and generation across various …

As Multimodal Large Language Models (MLLMs) gain widespread applicability, it is
becoming increasingly desirable to adapt them for diverse user needs. In this paper, we …