Yu-Chiang Frank Wang

Test-Time Alignment for Large Language Models via Textual Model Predictive Control

Guardrail-Agnostic Societal Bias Evaluation in Large Vision-Language Models

MV-SAM: Multi-view Promptable Segmentation using Pointmap Guidance

DLER: Doing Length pEnalty Right — Incentivizing More Intelligence per Token via Reinforcement Learning

OmniVinci: Enhancing Architecture and Data for Omni-Modal Understanding LLM

MoSEL: Modular Self-Reflective Learning for Embodied Decision-Making

Zoom-Zero: Reinforced Coarse-to-Fine Video Understanding via Temporal Zoom-in

Quantile Rendering: Efficiently Embedding High-dimensional Feature on 3D Gaussian Splatting

SAKE: Towards Editing Auditory Attribute Knowledge of Large Audio-Language Models

Learning Skills from Action-Free Videos

FormCraft: Beyond Documents - Benchmarking Form Intelligence

You Are What You Train: Rethinking Training Data Quality, Targets, and Architectures for Universal Speech Enhancement

Unified Reinforcement and Imitation Learning for Vision-Language Models

EMLoC: Emulator-based Memory-efficient Fine-tuning with LoRA Correction

SANER: Annotation-free Societal Attribute Neutralizer for Debiasing CLIP

ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planning

UniWav: Towards Unified Pre-training for Speech Representation Learning and Generation

Segment, Associate, and Classify: Decoupled Audio-Visual Segmentation Framework

Temporal Prompting Matters: Rethinking Referring Video Object Segmentation

VoiceNoNG: High-Quality Speech Editing Model without Hallucinations