Yuan Meng

Block-wise Adaptive Caching for Accelerating Diffusion Policy

SP-VLA: A Joint Model Scheduling and Token Pruning Approach for VLA Model Acceleration

Adaptive Mixture of Disentangled Experts for Dynamic Graph Out-of-Distribution Generalization

PhysUniBench: A Multi-Modal Physics Reasoning Benchmark at Undergraduate Level

Sparse ActionGen: Accelerating Diffusion Policy with Real-time Pruning

WiSparse: Boosting LLM Inference Efficiency with Weight-Aware Mixed Activation Sparsity

Point2Primitive: CAD Reconstruction from Point Cloud by Direct Primitive Prediction

Embodied Graph: A Relational World Model

Evaluating the Generalization Ability of Quantized LLMs: Benchmark, Analysis, and Toolbox

MindDETR: Beyond Semantics, Exploring Positional Cues from Brain Activity

EMS: Adaptive Evict-then-Merge Strategy for Head-wise KV Cache Compression Based on Global-Local Importance

GAQAT: Gradient-Adaptive Quantization-Aware Training for Domain Generalization

Accelerate Quantization Aware Training for Diffusion Models with Difficulty-aware Time Allocation

One QuantLLM for ALL: Fine-tuning Quantized LLMs Once for Efficient Deployments

3DAxisPrompt: Promoting the 3D Grounding and Reasoning in GPT-4o