Cho-Jui Hsieh

Self-Forcing++: Towards Minute-Scale High-Quality Video Generation

Concept or Skills? Rethinking Instruction Selection for Multi-modal Models

Fourier Neural Filter as Generic Vision Backbone

Exploring Expert Failures Improves LLM Agent Tuning

LLM-guided Hierarchical Retrieval

DialectGen: Benchmarking and Improving Dialect Robustness in Multimodal Generation

IRIS: Intrinsic Reward Image Synthesis

POME: Post Optimization Model Edit via Matrix Orthogonalization

When Distance Distracts: Representation Distance Bias in BT-Loss for Reward Models

VisualThinker: First ever R1-Zero's Aha Moment on just a 2B non-SFT Model

MAS-SAT: Synergizing ML-Assisted and Standalone Solvers for SAT Solving

SLIM: Structure-aware Low-rank Inference Model

Rethinking RL Evaluation: Can Benchmarks Truly Reveal Failures of RL Methods?

SeedFT: Structure-Preserving Fusion for Multi-Seed LLM Fine-Tuning

Latent Video Dataset Distillation

LoRA Done RITE: Robust Invariant Transformation Equilibration for LoRA Optimization

On the Loss of Context Awareness in General Instruction Fine-tuning

Unlabeled Data Improves Fine-Grained Image Zero-shot Classification with Multimodal LLMs

OR-Bench: An Over-Refusal Benchmark for Large Language Models

Large Language Models are Interpretable Learners

The Crystal Ball Hypothesis in diffusion models: Anticipating object positions from initial noise

Is Your Multimodal Language Model Oversensitive to Safe Queries?

Don’t Think Longer, Think Wisely: Optimizing Thinking Dynamics for Large Reasoning Models

Sparse MeZO: Less Parameters for Better Performance in Zeroth-Order LLM Fine-Tuning

Learn from the Past: Dynamic Data Pruning with Historically Weighted Bernoulli Sampling

Sparse MeZO: Less Parameters for Better Performance in Zeroth-Order LLM Fine-Tuning

SeedLoRA: A Fusion Approach to Efficient LLM Fine-Tuning

SeedLoRA: A Fusion Approach to Efficient LLM Fine-Tuning

Certified Training with Branch-and-Bound: A Case Study on Lyapunov-stable Neural Control

OR-Bench: An Over-Refusal Benchmark for Large Language Models

On the loss of context-awareness in general instruction finetuning

MuLan: Multimodal-LLM Agent for Progressive and Interactive Multi-Object Diffusion

Provably Noise-Resilient Training of Parameterized Quantum Circuits