16 citations · 37 across the 29 of their papers we have counts for
21 papers · 1 filter
GeoWAM: Visual Geometry World Action Models for Autonomous Driving
Yiren Lu, Xin Ye, Jiaming Liu +10
World action models (WAMs) have recently gained increasing attention as a framework for jointly modeling scene evolution and ego actions in autonomous driving. Most existing WAMs l…
Rethinking Classifier-Free Guidance in On-Policy Diffusion Distillation
Bingnan Li, Haozhe Wang, Haozhong Xiong +5
On-policy distillation (OPD) adapts diffusion models by querying a teacher along trajectories generated by the current student, but how it should behave under classifier-free guida…
CollectionLoRA: Collecting 50 Effects in 1 LoRA via Multi-Teacher On-Policy Distillation
Fangtai Wu, Hailong Guo, Shijie Huang +7
Customized image editing aims to equip pre-trained diffusion models with specific visual effects using limited paired data, typically via Low-Rank Adaptation (LoRA). As the number…
MoASE++: Mixture of Activation Sparsity Experts with Domain-Adaptive On-policy Distillation for Continual Test Time Adaptation
Ronyu Zhang, Aosong Cheng, Gaole Dai +8
Continual test-time adaptation adapts a source-pretrained model to non-stationary, unlabeled target streams while retaining past competence, yet texture-biased backbones risk error…
BEVUDA++: Geometric-aware Unsupervised Domain Adaptation for Multi-View 3D Object Detection
Rongyu Zhang, Jiaming Liu, Xiaoqi Li +5
Vision-centric Bird's Eye View (BEV) perception holds considerable promise for autonomous driving. Recent studies have prioritized efficiency or accuracy enhancements, yet the issu…
Lift3D Foundation Policy: Lifting 2D Large-Scale Pretrained Models for Robust 3D Robotic Manipulation
Yueru Jia, Jiaming Liu, Sixiang Chen +8
3D geometric information is essential for manipulation tasks, as robots need to perceive the 3D environment, reason about spatial relationships, and interact with intricate spatial…
MAVIS: Mathematical Visual Instruction Tuning with an Automatic Data Engine
Renrui Zhang, Xinyu Wei, Dongzhi Jiang +11
The mathematical capabilities of Multi-modal Large Language Models (MLLMs) remain under-explored with three areas to be improved: visual encoding of math diagrams, diagram-language…
MR-MLLM: Mutual Reinforcement of Multimodal Comprehension and Vision Perception
Guanqun Wang, Xinyu Wei, Jiaming Liu +5
In recent years, multimodal large language models (MLLMs) have shown remarkable capabilities in tasks like visual question answering and common sense reasoning, while visual percep…
Decomposing the Neurons: Activation Sparsity via Mixture of Experts for Continual Test Time Adaptation
Rongyu Zhang, Aosong Cheng, Yulin Luo +8
Continual Test-Time Adaptation (CTTA), which aims to adapt the pre-trained model to ever-evolving target domains, emerges as an important task for vision models. As current vision…
LLM as Dataset Analyst: Subpopulation Structure Discovery with Large Language Model
Yulin Luo, Ruichuan An, Bocheng Zou +3
The distribution of subpopulations is an important property hidden within a dataset. Uncovering and analyzing the subpopulation distribution within datasets provides a comprehensiv…
Any2Point: Empowering Any-modality Large Models for Efficient 3D Understanding
Yiwen Tang, Ray Zhang, Jiaming Liu +8
Large foundation models have recently emerged as a prominent focus of interest, attaining superior performance in widespread scenarios. Due to the scarcity of 3D data, many efforts…
Point-DETR3D: Leveraging Imagery Data with Spatial Point Prior for Weakly Semi-supervised 3D Object Detection
Hongzhi Gao, Zheng Chen, Zehui Chen +4
Training high-accuracy 3D detectors necessitates massive labeled 3D annotations with 7 degree-of-freedom, which is laborious and time-consuming. Therefore, the form of point annota…
A Vanilla Multi-Task Framework for Dense Visual Prediction Solution to 1st VCL Challenge -- Multi-Task Robustness Track
Zehui Chen, Qiuchen Wang, Zhenyu Li +3
In this report, we present our solution to the multi-task robustness track of the 1st Visual Continual Learning (VCL) Challenge at ICCV 2023 Workshop. We propose a vanilla framewor…
Efficient Deweather Mixture-of-Experts with Uncertainty-aware Feature-wise Linear Modulation
Rongyu Zhang, Yulin Luo, Jiaming Liu +8
The Mixture-of-Experts (MoE) approach has demonstrated outstanding scalability in multi-task learning including low-level upstream tasks such as concurrent removal of multiple adve…
Cloud-Device Collaborative Learning for Multimodal Large Language Models
Guanqun Wang, Jiaming Liu, Chenxuan Li +8
The burgeoning field of Multimodal Large Language Models (MLLMs) has exhibited remarkable performance in diverse tasks such as captioning, commonsense reasoning, and visual scene u…
LiDAR-LLM: Exploring the Potential of Large Language Models for 3D LiDAR Understanding
Senqiao Yang, Jiaming Liu, Ray Zhang +7
Recently, Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) have shown promise in instruction following and 2D image understanding. While these models are p…
Learning from Mistakes: Iterative Prompt Relabeling for Text-to-Image Diffusion Model Training
Xinyan Chen, Jiaxin Ge, Tianjun Zhang +2
Diffusion models have shown impressive performance in many domains. However, the model's capability to follow natural language instructions (e.g., spatial relationships between obj…
Continual-MAE: Adaptive Distribution Masked Autoencoders for Continual Test-Time Adaptation
Jiaming Liu, Ran Xu, Senqiao Yang +5
Continual Test-Time Adaptation (CTTA) is proposed to migrate a source pre-trained model to continually changing target distributions, addressing real-world dynamism. Existing CTTA…
MoSA: Mixture of Sparse Adapters for Visual Efficient Tuning
Qizhe Zhang, Bocheng Zou, Ruichuan An +2
With the rapid growth in the scale of pre-trained foundation models, parameter-efficient fine-tuning techniques have gained significant attention, among which Adapter Tuning is the…
FreeKD: Knowledge Distillation via Semantic Frequency Prompt
Yuan Zhang, Tao Huang, Jiaming Liu +3
Knowledge distillation (KD) has been applied to various tasks successfully, and mainstream methods typically boost the student model via spatial imitation losses. However, the cons…
Biphasic Face Photo-Sketch Synthesis via Semantic-Driven Generative Adversarial Network with Graph Representation Learning
Xingqun Qi, Muyi Sun, Zijian Wang +5
Biphasic face photo-sketch synthesis has significant practical value in wide-ranging fields such as digital entertainment and law enforcement. Previous approaches directly generate…