7 papers
Test-Time Registers as Global Priors for Tokenized Image Generation
Cheng-Yao Hong, Yifan Wang, Yuewei Lin +1
Attention-based models often develop attention sinks, where a small number of tokens repeatedly attract attention and accumulate unusually large activations. In vision transformers…
VisReason: A Large-Scale Dataset for Visual Chain-of-Thought Reasoning
Lingxiao Li, Yifan Wang, Xinyan Gao +3
Chain-of-Thought (CoT) prompting has proven remarkably effective for eliciting complex reasoning in large language models (LLMs). Yet, its potential in multimodal large language mo…
Ouroboros: Single-step Diffusion Models for Cycle-consistent Forward and Inverse Rendering
Shanlin Sun, Yifan Wang, Hanwen Zhang +5
While multi-step diffusion models have advanced both forward and inverse rendering, existing approaches often treat these problems independently, leading to cycle inconsistency and…
Route Experts by Sequence, not by Token
Tiansheng Wen, Yifei Wang, Aosong Feng +7
Mixture-of-Experts (MoE) architectures scale large language models (LLMs) by activating only a subset of experts per token, but the standard TopK routing assigns the same fixed num…
CSRv2: Unlocking Ultra-Sparse Embeddings
Lixuan Guo, Yifei Wang, Tiansheng Wen +5
In the era of large foundation models, the quality of embeddings has become a central determinant of downstream task performance and overall system capability. Yet widely used dens…
HorizonForge: Driving Scene Editing with Any Trajectories and Any Vehicles
Yifan Wang, Francesco Pittaluga, Zaid Tasneem +3
Controllable driving scene generation is critical for realistic and scalable autonomous driving simulation, yet existing approaches struggle to jointly achieve photorealism and pre…