Chun Yuan

OptMerge: Unifying Multimodal LLM Capabilities and Modalities via Model Merging

OmniPortrait: Fine-Grained Personalized Portrait Synthesis via Pivotal Optimization

AlphaVAE: Unified End-to-End RGBA Image Reconstruction and Generation with Alpha-Aware Representation Learning

PromptHub: Enhancing Multi-Prompt Visual In-Context Learning with Locality-Aware Fusion, Concentration and Alignment

MIGA: Mutual Information-Guided Attack on Denoising Models for Semantic Manipulation

Enhancing Geometric Perception in VLMs via Translator-Guided Reinforcement Learning

FlashVSR: Towards Real-time Diffusion-Based Streaming Video Super-Resolution

Mixture of Neuron Experts

UltraHorizon: Benchmarking LLM-Agent Capabilities in Ultra Long-Horizon Scenarios

Benchmarking and Advancing Quantization-Aware Training for Reasoning Models

InstructEngine: Instruction-driven Text-to-Image Alignment

DiffuSpec: Unlocking Diffusion Language Models for Speculative Decoding

RaBEL: Scale-Aware Radial-Basis Embeddings for Tabular Foundation Models

Learn to Learn Consistently via Meta Self-distillation for Few-shot Classification

Open-Vocabulary Customization from CLIP via Data-Free Knowledge Distillation

Quantization Hurts Reasoning? An Empirical Study on Quantized Reasoning Models

FlatQuant: Flatness Matters for LLM Quantization

SCOUT: Teaching Pre-trained Language Models to Enhance Reasoning via Flow Chain-of-Thought

Towards Robust Uncertainty Calibration for Composed Image Retrieval

Towards Implicit Aggregation: Robust Image Representation for Place Recognition in the Transformer Era

Safe-Sora: Safe Text-to-Video Generation via Graphical Watermarking

ChartMoE: Mixture of Diversely Aligned Expert Connector for Chart Understanding

NaDRO: Leveraging Dual-Reward Strategies for LLMs Training on Noisy Data

Whoever Started the interference Should End It: Guiding Data-Free Model Merging via Task Vectors

Enhancing Logits Distillation with Plug&Play Kendall's $\tau$ Ranking Loss

Preference Optimization for Combinatorial Optimization Problems

Enhancing Logits Distillation with Plug&Play Kendall's $\tau$ Ranking Loss

A Simple Linear Patch Revives Layer-Pruned Large Language Models

ChartBench: A Benchmark for Complex Visual Reasoning in Charts

IMDPrompter: Adapting SAM to Image Manipulation Detection by Cross-View Automated Prompt Learning

Preference Optimization for Combinatorial Optimization Problems

Modeling Multi-Task Model Merging as Adaptive Projective Gradient Descent

FlatQuant: Flatness Matters for LLM Quantization

LoRA Recycle: Towards Fine-Tuning-Free Visual Foundation Model via Double-Efficient Data-Free Meta-Learning

LEARN TO LEARN CONSISTENTLY

Unifying All Species: LLM-based Hyper-Heuristics for Multi-objective Optimization