3 accepted papers
Improving the reasoning capabilities of diffusion-based large language models (dLLMs) through reinforcement learning (RL) remains an open problem. The intractability of dLLMs likelihood function necessitates approximating the current, old, and reference policy likelihoods at each policy optimization…
Subword tokenization methods like Byte Pair Encoding (BPE) are widely used in large language models due to their balance of vocabulary compactness and representational power. However, they suffer from inefficiencies in representing rare words and require large embedding matrices. Character-level mod
We address the problem of incremental sequence classification, where predictions are updated as new elements in the sequence are revealed. Drawing on temporal-difference learning from reinforcement learning, we identify a temporal-consistency condition that successive predictions should satisfy. We…