This paper introduces a novel generalized self-imitation learning ($\textbf {GSIL} $)
framework, which effectively and efficiently aligns large language models with offline …

Confidence calibration in LLMs, ie, aligning their self-assessed confidence with the actual
accuracy of their responses, enabling them to self-evaluate the correctness of their outputs …

Existing preference optimization objectives for language model alignment require additional
hyperparameters that must be extensively tuned to achieve optimal performance, increasing …

Pretraining molecular representations is crucial for drug and material discovery. Recent
methods focus on learning representations from geometric structures, effectively capturing …

Test-Time Adaptation (TTA) has emerged as a promising paradigm for enhancing the
generalizability of models. However, existing mainstream TTA methods, predominantly …