Many modern language models have adopted hybrid architectures that incorporate both linear-time recurrence and softmax attention. Modern recurrence layers provide a highly expressive mixing mechanism while ensuring high throughput by compressing contextual information into a fixed-size state, but underperform softmax a...
Xing-Yu Shen, Yingfa Chen, Zi-Qing Qiao et al.· 0 citations
Language model agents increasingly automate post-training, but when given the full training stack they converge on supervised fine-tuning and spend most effort revising training data. Existing systems give the agent one kind of control, selecting rows from a fixed pool or calling a generator, whereas a human data engin...
Bing-Xiang He, Peng Sun, Shi-Hao Peng et al.· 0 citations
Downstream RLHF experiments additionally show that using DRM as the training-time reward leads to improved policy performance, directly validating the practical benefit of diffusion-based reward modeling for RLHF training.
Xiang-Yang Wang, Bing-Xiang He, Ze-Yuan Liu et al.· arXiv.org· 0 citations
StudyBench is introduced, a controlled physics benchmark that directly measures how efficiently a self-evolution method converts training material into capability, and turns self-evolution progress from an open-ended pursuit into a measurable target for future research.
Ying-Hao Chen, Zi-Xi Chen, Bingxiang He et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.