Ying Nian Wu

Score Distillation Beyond Acceleration: Generative Modeling from Corrupted Data

Beyond Pass@ 1: Self-Play with Variational Problem Synthesis Sustains RLVR

MILR: Improving Multimodal Image Generation via Test-Time Latent Reasoning

Reasoning via Test-Time Instance-Level Policy Gradient in Latent Space

Generative Actor Critic

EdiVal-Agent: An Object-Centric Framework for Automated, Fine-Grained Evaluation of Multi-Turn Editing

Fourier Neural Filter as Generic Vision Backbone

Learning to Rank Chain-of-Thought: Using a Small Model

Continuous-Time Discrete Markov Bridge

GTD: Dynamic Generation of Multi LLM Agents Communication Topologies with Graph Diffusion Models

``Noisier'’ Noise Contrastive Estimation is (Almost) Maximum Likelihood

Natural gradient Bayesian sampling automatically emerges in canonical cortical circuits

Multivariate Time Series Forecasting with Fourier Neural Filter

Distortion Regularization for Distance-Preserving Graph Embedding

Quantifying QiGan: What Shape Patterns Does a DNN Exploit in Go?

You Are What Role You Play: Directing AI Values Through Role Assignment​

Two Lenses are Better Than One: Dual Vector Quantization for Self-Supervised Graph Learning

Multimodal Function Vectors for Spatial Relations

Energy-Driven Steering: Reducing False Refusals in Large Language Models

Latent Adaptation with Masked Policy for Diffusion Language Models

Sandbox-RL: Scalable Multi-LLMs Optimization through Sandbox-Based Reinforcement Learning

Relational Knowledge Distillation Using Finetuned Function Vectors

FFT-Accelerated Auxiliary Variable MCMC for Fermionic Lattice Models: A Determinant-Free Approach with $O(N \log N)$ Complexity

On Conformal Isometry of Grid Cells: Learning Distance-Preserving Position Embedding

Place Cells as Multi-Scale Position Embeddings: Random Walk Transition Kernels for Path Planning

SlowFast-VGen: Slow-Fast Learning for Action-Driven Long Video Generation

Optimizing the Unknown: Black Box Bayesian Optimization with Energy-Based Model and Reinforcement Learning

MLZero: A Multi-Agent System for End-to-end Machine Learning Automation

Visual Agents as Fast and Slow Thinkers

Diff-PIC: Revolutionizing Particle-In-Cell Nuclear Fusion Simulation with Diffusion Models

SwS: Self-aware Weakness-driven Problem Synthesis in Reinforcement Learning for LLM Reasoning

DS-LLM: Leveraging Dynamical Systems to Enhance Both Training and Inference of Large Language Models

An Expressive and Self-Adaptive Dynamical System for Efficient Function Learning

Latent Thought Models with Variational Bayes Inference-Time Computation

Inertial Confinement Fusion Forecasting via Large Language Models

Think Twice Before You Act: Improving Inverse Problem Solving With MCMC

Disentangling the QiGan Encoded by a DNN towards the Go Game

A Brain-Inspired Machine Learning Paradigm for Nature-Powered Equation Solving

Conditional Trajectories in Diffusion Models - Modeling Galaxy Evolution from Redshift

TeamCraft: A Benchmark for Embodied Multi-Agent Systems in Minecraft