8 papers
Prism: Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models
Jinbin Bai, Yixuan Li, Yuchen Zhu +8
Inference-time compute has re-emerged as a practical way to improve LLM reasoning. Most test-time scaling (TTS) algorithms rely on autoregressive decoding, which is ill-suited to d…
Towards Interpretability Without Sacrifice: Faithful Dense Layer Decomposition with Mixture of Decoders
James Oldfield, Shawn Im, Sharon Li +3
Multilayer perceptrons (MLPs) are an integral part of large language models, yet their dense representations render them difficult to understand, edit, and steer. Recent methods le…
A Unified Understanding and Evaluation of Steering Methods
Shawn Im, Sharon Li
Latent space steering methods provide a practical approach to controlling large language models by applying steering vectors to intermediate activations, guiding outputs toward des…
FluencyVE: Marrying Temporal-Aware Mamba with Bypass Attention for Video Editing
Mingshu Cai, Yixuan Li, Osamu Yoshie +1
Large-scale text-to-image diffusion models have achieved unprecedented success in image generation and editing. However, extending this success to video editing remains challenging…
Visual Instruction Bottleneck Tuning
Changdae Oh, Jiatong Li, Shawn Im +1
Despite widespread adoption, multimodal large language models (MLLMs) suffer performance degradation when encountering unfamiliar queries under distribution shifts. Existing method…
STAR: Speech-to-Audio Generation via Representation Learning
Zeyu Xie, Xuenan Xu, Yixuan Li +2
This work presents STAR, the first end-to-end speech-to-audio generation framework, designed to enhance efficiency and address error propagation inherent in cascaded systems. Unlik…