Jiwen Lu

29

Papers

176

Total Citations

Papers (29)

MADTP: Multimodal Alignment-Guided Dynamic Token Pruning for Accelerating Vision-Language Transformer

FlowIE: Efficient Image Enhancement via Rectified Flow

Efficient Inference of Vision Instruction-Following Models with Elastic Cache

ThinkBot: Embodied Instruction Following with Thought Chain Reasoning

DPMesh: Exploiting Diffusion Prior for Occluded Human Mesh Recovery

EmbodiedOcc: Embodied 3D Occupancy Prediction for Vision-based Online Scene Understanding

Once for Both: Single Stage of Importance and Sparsity Search for Vision Transformer Compression

UniPre3D: Unified Pre-training of 3D Point Cloud Models with Cross-Modal Gaussian Splatting

Path Choice Matters for Clear Attributions in Path Methods

Learning Dual-Level Deformable Implicit Representation for Real-World Scale Arbitrary Super-Resolution

SparseMM: Head Sparsity Emerges from Visual Concept Responses in MLLMs

D3QE: Learning Discrete Distribution Discrepancy-aware Quantization Error for Autoregressive-Generated Image Detection

PlaneRAS: Learning Planar Primitives for 3D Plane Recovery

Text-guided Sparse Voxel Pruning for Efficient 3D Visual Grounding

IGL-Nav: Incremental 3D Gaussian Localization for Image-goal Navigation

Learning Counterfactually Decoupled Attention for Open-World Model Attribution

X-3D: Explicit 3D Structure Modeling for Point Cloud Recognition

LowRankOcc: Tensor Decomposition and Low-Rank Recovery for Vision-based 3D Semantic Occupancy Prediction

SelfOcc: Self-Supervised Vision-Based 3D Occupancy Prediction

MirageRoom: 3D Scene Segmentation with 2D Pre-trained Models by Mirage Projection

GaussianFormer-2: Probabilistic Gaussian Superposition for Efficient 3D Occupancy Prediction

UniGoal: Towards Universal Zero-shot Goal-oriented Navigation

GaussianWorld: Gaussian World Model for Streaming 3D Occupancy Prediction

Memory-based Adapters for Online 3D Scene Perception

Towards Accurate Post-training Quantization for Diffusion Models

Segment and Caption Anything

EfficientLLaVA: Generalizable Auto-Pruning for Large Vision-language Models

Narrative Action Evaluation with Prompt-Guided Multimodal Interaction

SpectralAR: Spectral Autoregressive Visual Generation