Ge Li

33

Papers

140

Total Citations

Papers (33)

ST-LLM: Large Language Models Are Effective Temporal Learners

Open the Black Box: Step-based Policy Updates for Temporally-Correlated Episodic Reinforcement Learning

Flow4Agent: Long-form Video Understanding via Motion Prior from Optical Flow

ContextHOI: Spatial Context Learning for Human-Object Interaction Detection

Hot or Cold? Adaptive Temperature Sampling for Code Generation with Large Language Models

BT-Adapter: Video Conversation is Feasible Without Video Instruction Tuning

Learning Background Prompts to Discover Implicit Knowledge for Open Vocabulary Object Detection

Dataflow-Guided Neuro-Symbolic Language Models for Type Inference

Graph Convolutional Label Noise Cleaner: Train a Plug-And-Play Action Classifier for Anomaly Detection

Deep Image Spatial Transformation for Person Image Generation

Neural Texture Extraction and Distribution for Controllable Person Image Synthesis

Self-Supervised Arbitrary-Scale Point Clouds Upsampling via Implicit Neural Representation

Contextual Debiasing for Visual Recognition With Causal Mechanisms

Revisiting Temporal Modeling for CLIP-Based Image-to-Video Knowledge Transferring

Improving Graph Representation for Point Cloud Segmentation via Attentive Filtering

Efficient Hierarchical Entropy Model for Learned Point Cloud Compression

StructureFlow: Image Inpainting via Structure-Aware Appearance Flow

AttPool: Towards Hierarchical Feature Representation in Graph Convolutional Networks via Attention Mechanism

PIRenderer: Controllable Portrait Image Generation via Semantic Neural Rendering

Structure-Transformed Texture-Enhanced Network for Person Image Synthesis

AdaNIC: Towards Practical Neural Image Compression via Dynamic Transform Routing

Salient Object Detection for Point Clouds

DGTalker: Disentangled Generative Latent Space Learning for Audio-Driven Gaussian Talking Heads

Orchestrating the Symphony of Prompt Distribution Learning for Human-Object Interaction Detection

Point Cloud Semantic Segmentation with Sparse and Inhomogeneous Annotations

MUSE: Mamba Is Efficient Multi-scale Learner for Text-video Retrieval

Multi-mapping Image-to-Image Translation via Learning Disentanglement

Code Generation as a Dual Task of Code Summarization

Integrating Tree Path in Transformer for Code Representation

SKFlow: Learning Optical Flow with Super Kernels

Learning to Share in Networked Multi-Agent Reinforcement Learning

Fine-Tuning Pre-Trained Language Models Effectively by Optimizing Subnetworks Adaptively

Birder: Communication-Efficient 1-bit Adaptive Optimizer for Practical Distributed DNN Training