15 citations · 51 across the 26 of their papers we have counts for
28 papers
DecoMoE: Decoupling Visual Propagation and Expert Computation for Efficient Multimodal MoE Inference
Xudong Tan, Peng Ye, Ming Xie +4
Multimodal mixture-of-experts (MoE) models combine sparse expert activation with visual-language capabilities, yet their inference remains costly because long visual-token sequence…
CTTS: Collective Test-Time Scaling
Zhende Song, Shengji Tang, Peng Ye +4
Test-time scaling (TTS) has emerged as a promising, training-free approach for enhancing large language model (LLM) performance. However, the efficacy of existing methods, such as…
Decouple and Orthogonalize: A Data-Free Framework for LoRA Merging
Shenghe Zheng, Hongzhi Wang, Chenyu Huang +5
With more open-source models available for diverse tasks, model merging has gained attention by combining models into one, reducing training, storage, and inference costs. Current…
Dynamic Base model Shift for Delta Compression
Chenyu Huang, Peng Ye, Shenghe Zheng +4
Transformer-based models with the pretrain-finetune paradigm bring about significant progress, along with the heavy storage and deployment costs of finetuned models on multiple tas…
Breaking the Compression Ceiling: Data-Free Pipeline for Ultra-Efficient Delta Compression
Xiaohui Wang, Peng Ye, Chenyu Huang +5
With the rise of the fine-tuned-pretrained paradigm, storing numerous fine-tuned models for multi-tasking creates significant storage overhead. Delta compression alleviates this by…
Seeing Delta Parameters as JPEG Images: Data-Free Delta Compression with Discrete Cosine Transform
Chenyu Huang, Peng Ye, Xiaohui Wang +5
With transformer-based models and the pretrain-finetune paradigm becoming mainstream, the high storage and deployment costs of individual finetuned models on multiple tasks pose cr…
DeRS: Towards Extremely Efficient Upcycled Mixture-of-Experts Models
Yongqi Huang, Peng Ye, Chenyu Huang +5
Upcycled Mixture-of-Experts (MoE) models have shown great potential in various tasks by converting the original Feed-Forward Network (FFN) layers in pre-trained dense models into M…
DualMamba: A Lightweight Spectral-Spatial Mamba-Convolution Network for Hyperspectral Image Classification
Jiamu Sheng, Jingyi Zhou, Jiong Wang +2
The effectiveness and efficiency of modeling complex spectral-spatial relations are both crucial for Hyperspectral image (HSI) classification. Most existing methods based on CNNs a…
Adapter-X: A Novel General Parameter-Efficient Fine-Tuning Framework for Vision
Minglei Li, Peng Ye, Yongqi Huang +5
Parameter-efficient fine-tuning (PEFT) has become increasingly important as foundation models continue to grow in both popularity and size. Adapter has been particularly well-recei…
EMR-Merging: Tuning-Free High-Performance Model Merging
Chenyu Huang, Peng Ye, Tao Chen +3
The success of pretrain-finetune paradigm brings about the release of numerous model weights. In this case, merging models finetuned on different tasks to enable a single model wit…
ClipSAM: CLIP and SAM Collaboration for Zero-Shot Anomaly Segmentation
Shengze Li, Jianjian Cao, Peng Ye +3
Recently, foundational models such as CLIP and SAM have shown promising performance for the task of Zero-Shot Anomaly Segmentation (ZSAS). However, either CLIP-based or SAM-based Z…
Merging Vision Transformers from Different Tasks and Domains
Peng Ye, Chenyu Huang, Mingzhu Shen +4
This work targets to merge various Vision Transformers (ViTs) trained on different tasks (i.e., datasets with different object categories) or domains (i.e., datasets with the same…
Partial Fine-Tuning: A Successor to Full Fine-Tuning for Vision Transformers
Peng Ye, Yongqi Huang, Chongjun Tu +4
Fine-tuning pre-trained foundation models has gained significant popularity in various research fields. Existing methods for fine-tuning can be roughly divided into two categories,…
Efficient Architecture Search via Bi-level Data Pruning
Chongjun Tu, Peng Ye, Weihao Lin +5
Improving the efficiency of Neural Architecture Search (NAS) is a challenging but significant task that has received much attention. Previous works mainly adopted the Differentiabl…
BridgeNet: Comprehensive and Effective Feature Interactions via Bridge Feature for Multi-task Dense Predictions
Jingdong Zhang, Jiayuan Fan, Peng Ye +5
Multi-task dense prediction aims at handling multiple pixel-wise prediction tasks within a unified network simultaneously for visual scene understanding. However, cross-task featur…
Towards an end-to-end artificial intelligence driven global weather forecasting system
Kun Chen, Lei Bai, Fenghua Ling +9
The weather forecasting system is important for science and society, and significant achievements have been made in applying artificial intelligence (AI) to medium-range weather fo…
Experts Weights Averaging: A New General Training Scheme for Vision Transformers
Yongqi Huang, Peng Ye, Xiaoshui Huang +4
Structural re-parameterization is a general training scheme for Convolutional Neural Networks (CNNs), which achieves performance improvement without increasing inference cost. As V…
Boosting Residual Networks with Group Knowledge
Shengji Tang, Peng Ye, Baopu Li +5
Recent research understands the residual networks from a new perspective of the implicit ensemble model. From this view, previous methods such as stochastic depth and stimulative t…
Exploring Multi-Timestep Multi-Stage Diffusion Features for Hyperspectral Image Classification
Jingyi Zhou, Jiamu Sheng, Jiayuan Fan +4
The effectiveness of spectral-spatial feature learning is crucial for the hyperspectral image (HSI) classification task. Diffusion models, as a new class of groundbreaking generati…
Stimulative Training++: Go Beyond The Performance Limits of Residual Networks
Peng Ye, Tong He, Shengji Tang +4
Residual networks have shown great success and become indispensable in recent deep neural network models. In this work, we aim to re-investigate the training process of residual ne…
A2S-NAS: Asymmetric Spectral-Spatial Neural Architecture Search For Hyperspectral Image Classification
Lin Zhan, Jiayuan Fan, Peng Ye +1
Existing deep learning-based hyperspectral image (HSI) classification works still suffer from the limitation of the fixed-sized receptive field, leading to difficulties in distinct…
-DARTS++: Bi-level Regularization for Proxy-robust Differentiable Architecture Search
Peng Ye, Tong He, Baopu Li +3
Neural Architecture Search has attracted increasing attention in recent years. Among them, differential NAS approaches such as DARTS, have gained popularity for the search efficien…
Stimulative Training of Residual Networks: A Social Psychology Perspective of Loafing
Peng Ye, Shengji Tang, Baopu Li +2
Residual networks have shown great success and become indispensable in today's deep models. In this work, we aim to re-investigate the training process of residual networks from a…
Efficient Joint-Dimensional Search with Solution Space Regularization for Real-Time Semantic Segmentation
Peng Ye, Baopu Li, Tao Chen +6
Semantic segmentation is a popular research topic in computer vision, and many efforts have been made on it with impressive results. In this paper, we intend to search an optimal n…