3 papers
cs.LG2026
GuidanceNFT: NFT-Based Reward Fine-Tuning of Guided Diffusion Models
Jiayin Liu, Zeyu Michael Li, Bingshuo Qian +1
DiffusionNFT (NFT) fine-tunes diffusion models toward task-specific rewards, but optimizing these rewards does not guarantee fidelity to the full prompt. Although classifier-free g…
cs.CL2026
ELF-REG: Scaling Continuous Diffusion Language Models to Reasoning Tasks
Zeyu Michael Li, William Xingxu Chen, Bingshuo Qian +2
Fully continuous diffusion language models (dLMs) denoise continuous representations without intermediate discretization, then decode all response tokens in parallel at the final s…
cs.CL2024
Controlled Low-Rank Adaptation with Subspace Regularization for Continued Training on Large Language Models
Yuheng Lu, Bingshuo Qian, Caixia Yuan +2
Large language models (LLMs) exhibit remarkable capabilities in natural language processing but face catastrophic forgetting when learning new tasks, where adaptation to a new doma…