3 papers
cs.LG2026
AdaFlash: Adaptive Speculative Decoding via On-Policy Distilled Diffusion Drafters
Yu-Yang Qian, Hao-Cong Wu, Chen Chen +4
Speculative decoding, in which a lightweight draft model first generates a draft sequence that is then verified in parallel by the target model, has become a prevalent paradigm for…
cs.LG2026
The Structural Origin of Attention Sink: Variance Discrepancy, Super Neurons, and Dimension Disparity
Siquan Li, Kaiqi Jiang, Jiacheng Sun +1
Despite the prevalence of the attention sink phenomenon in Large Language Models (LLMs), where initial tokens disproportionately monopolize attention scores, its structural origins…
cs.CV2025
Towards Transformer-Based Aligned Generation with Self-Coherence Guidance
Shulei Wang, Wang Lin, Hai Huang +8
We introduce a novel, training-free approach for enhancing alignment in Transformer-based Text-Guided Diffusion Models (TGDMs). Existing TGDMs often struggle to generate semantical…