4 papers
OPTD: On-Policy Transition Distillation with Consistency-Guided Adaptive Compression for Few-Step Diffusion Language Models
Xiaocheng Lu, Hualei Zhang, Shuhan Guo +8
Diffusion language models (dLLMs) can predict many tokens in parallel, but accurate generation still requires many iterative denoising steps. Few-step distillation accelerates deco…
Mitigating Bias in Low-SNR Financial Reinforcement Learning via Quantum Representations
Zeyu Liu, Xuanzhi Feng, Sing Kwong Lai +6
The financial market is a typical low signal-to-noise ratio (SNR) setting, which often destabilizes off-policy maximum-entropy methods like Soft Actor-Critic (SAC). Specifically, n…
VisionCreator-R1: A Reflection-Enhanced Native Visual-Generation Agentic Model
Jinxiang Lai, Wenzhe Zhao, Zexin Lu +7
Visual content generation has advanced from single-image to multi-image workflows, yet existing agents remain largely plan-driven and lack systematic reflection mechanisms to corre…
LazyEviction: Lagged KV Eviction with Attention Pattern Observation for Efficient Long Reasoning
Haoyue Zhang, Hualei Zhang, Xiaosong Ma +2
Large Language Models (LLMs) exhibit enhanced capabilities by Chain-of-Thought reasoning. However, the extended reasoning sequences introduce significant GPU memory overhead due to…