no code implementations • 3 Feb 2025 • YuHang Zhou, Giannis Karamanolakis, Victor Soto, Anna Rumshisky, Mayank Kulkarni, Furong Huang, Wei Ai, Jianhua Lu
The recent success of specialized Large Language Models (LLMs) in domains such as mathematical reasoning and coding has led to growing interest in methods for merging these expert LLMs into a unified Mixture-of-Experts (MoE) model, with the goal of enhancing performance in each domain while retaining effectiveness on general tasks.
1 code implementation • 18 Nov 2024 • Shengchao Hu, YuHang Zhou, Ziqing Fan, Jifeng Hu, Li Shen, Ya zhang, DaCheng Tao
Training a generalizable agent to continually learn a sequence of tasks from offline trajectories is a natural requirement for long-lived agents, yet remains a significant challenge for current offline reinforcement learning (RL) algorithms.
1 code implementation • 2 Nov 2024 • Ziqing Fan, Shengchao Hu, YuHang Zhou, Li Shen, Ya zhang, Yanfeng Wang, DaCheng Tao
The purpose of offline multi-task reinforcement learning (MTRL) is to develop a unified policy applicable to diverse tasks without the need for online environmental interaction.
no code implementations • 18 Oct 2024 • Zhibin Wang, Shipeng Li, YuHang Zhou, Xue Li, Rong Gu, Nguyen Cam-Tu, Chen Tian, Sheng Zhong
In this paper, we revisit SLO and goodput metrics in LLM serving and propose a unified metric framework smooth goodput including SLOs and goodput to reflect the nature of user experience in LLM serving.
2 code implementations • 29 Sep 2024 • Haolin Li, YuHang Zhou, Ziheng Zhao, Siyuan Du, Jiangchao Yao, Weidi Xie, Ya zhang, Yanfeng Wang
To accomplish the above objective, we propose a novel framework named Low-Rank Knowledge Decomposition (LoRKD), which explicitly separates gradients from different tasks by incorporating low-rank expert modules and efficient knowledge separation convolution.
3D Medical Imaging Segmentation
Medical Image Classification
1 code implementation • 23 Aug 2024 • Xiaoyu Liu, Jiaxin Yuan, YuHang Zhou, Jingling Li, Furong Huang, Wei Ai
The essence of sequential recommender systems (RecSys) lies in understanding how users make decisions.
no code implementations • 9 Jul 2024 • YuHang Zhou, Siyuan Du, Haolin Li, Jiangchao Yao, Ya zhang, Yanfeng Wang
However, due to the gap between pre-training tasks (or modalities) and downstream tasks (or modalities), the real-world computation and speed constraints, it might not be straightforward to apply medical foundation models in the downstream scenarios.
1 code implementation • 24 Jun 2024 • Jing Zhu, YuHang Zhou, Shengyi Qian, Zhongmou He, Tong Zhao, Neil Shah, Danai Koutra
Associating unstructured data with structured information is crucial for real-world tasks that require relevance search.
1 code implementation • 19 Jun 2024 • YuHang Zhou, Jing Zhu, Paiheng Xu, Xiaoyu Liu, Xiyao Wang, Danai Koutra, Wei Ai, Furong Huang
Large language models (LLMs) have significantly advanced various natural language processing tasks, but deploying them remains computationally expensive.
1 code implementation • 14 Jun 2024 • YuHang Zhou, Zihua Zhao, Haolin Li, Siyuan Du, Jiangchao Yao, Ya zhang, Yanfeng Wang
Training a unified model to take multiple targets into account is a trend towards artificial general intelligence.
no code implementations • 8 Jun 2024 • YuHang Zhou, Wei Ai
The first signal is the student's self-consistency (consistency of student multiple outputs), which is a proxy of the student's confidence.
2 code implementations • 24 May 2024 • Xiyao Wang, Jiuhai Chen, Zhaoyang Wang, YuHang Zhou, Yiyang Zhou, Huaxiu Yao, Tianyi Zhou, Tom Goldstein, Parminder Bhatia, Furong Huang, Cao Xiao
In this paper, we propose SIMA, a framework that enhances visual and language modality alignment through self-improvement, eliminating the needs for external models or data.
Ranked #189 on
Visual Question Answering
on MM-Vet
1 code implementation • CVPR 2024 • YuHang Zhou, Haolin Li, Siyuan Du, Jiangchao Yao, Ya zhang, Yanfeng Wang
The popularity of large-scale pre-training has promoted the development of medical foundation models.
no code implementations • 7 Apr 2024 • Libo Qin, Qiguang Chen, YuHang Zhou, Zhi Chen, Yinghui Li, Lizi Liao, Min Li, Wanxiang Che, Philip S. Yu
To this end, in this paper, we present a thorough review and provide a unified perspective to summarize the recent progress as well as emerging trends in multilingual large language models (MLLMs) literature.
no code implementations • 7 Apr 2024 • YuHang Zhou, Zeping Li, Siyu Tian, Yuchen Ni, Sen Liu, Guangnan Ye, Hongfeng Chai
Large language models (LLMs) are increasingly being applied across various specialized fields, leveraging their extensive knowledge to empower a multitude of scenarios within these domains.
no code implementations • 4 Apr 2024 • Lei Zhang, YuHang Zhou, Yi Yang, Xinbo Gao
Despite providing high-performance solutions for computer vision tasks, the deep neural network (DNN) model has been proved to be extremely vulnerable to adversarial attacks.
no code implementations • CVPR 2024 • YuHang Zhou, Zhongyun Hua
In this paper, we discuss for the first time the concept of continual adversarial defense under a sequence of attacks, and propose a lifelong defense baseline called Anisotropic \& Isotropic Replay (AIR), which offers three advantages: (1) Isotropic replay ensures model consistency in the neighborhood distribution of new data, indirectly aligning the output preference between old and new tasks.
no code implementations • 14 Mar 2024 • Xiaoyu Liu, Paiheng Xu, Junda Wu, Jiaxin Yuan, Yifan Yang, YuHang Zhou, Fuxiao Liu, Tianrui Guan, Haoliang Wang, Tong Yu, Julian McAuley, Wei Ai, Furong Huang
Causal inference has shown potential in enhancing the predictive accuracy, fairness, robustness, and explainability of Natural Language Processing (NLP) models by capturing causal relationships among variables.
no code implementations • 22 Feb 2024 • YuHang Zhou, Xuan Lu, Wei Ai
In the rapidly evolving landscape of social media, the introduction of new emojis in Unicode release versions presents a structured opportunity to explore digital language evolution.
no code implementations • 20 Feb 2024 • YuHang Zhou, Yuchen Ni, Yunhui Gan, Zhangyue Yin, Xiang Liu, Jian Zhang, Sen Liu, Xipeng Qiu, Guangnan Ye, Hongfeng Chai
Results show varying degrees of financial irrationality among models, influenced by their design and training.
no code implementations • 1 Feb 2024 • Xiaowei Fu, YuHang Zhou, Lina Ma, Lei Zhang
Based on this finding, a Pixel Surgery and Semantic Regeneration (PSSR) model following the targeted therapy mechanism is developed, which has three merits: 1) To remove the salient attack, a score-based Pixel Surgery module is proposed, which retains the trivial attack as a kind of invariance information.
no code implementations • 22 Jan 2024 • YuHang Zhou, Paiheng Xu, Xiyao Wang, Xuan Lu, Ge Gao, Wei Ai
Our objective is to validate the hypothesis that ChatGPT can serve as a viable alternative to human annotators in emoji research and that its ability to explain emoji meanings can enhance clarity and transparency in online communications.
1 code implementation • 19 Jan 2024 • Xiyao Wang, YuHang Zhou, Xiaoyu Liu, Hongjin Lu, Yuancheng Xu, Feihong He, Jaehong Yoon, Taixi Lu, Gedas Bertasius, Mohit Bansal, Huaxiu Yao, Furong Huang
However, current MLLM benchmarks are predominantly designed to evaluate reasoning based on static information about a single image, and the ability of modern MLLMs to extrapolate from image sequences, which is essential for understanding our ever-changing world, has been less investigated.
1 code implementation • 15 Nov 2023 • YuHang Zhou, Paiheng Xu, Xiaoyu Liu, Bang An, Wei Ai, Furong Huang
We find that LMs, when encountering spurious correlations between a concept and a label in training or prompts, resort to shortcuts for predictions.
1 code implementation • 3 Nov 2023 • YuHang Zhou, Yu He, Siyu Tian, Yuchen Ni, Zhangyue Yin, Xiang Liu, Chuanjun Ji, Sen Liu, Xipeng Qiu, Guangnan Ye, Hongfeng Chai
While current tasks of converting natural language to SQL (NL2SQL) using Foundation Models have shown impressive achievements, adapting these approaches for converting natural language to Graph Query Language (NL2GQL) encounters hurdles due to the distinct nature of GQL compared to SQL, alongside the diverse forms of GQL.
no code implementations • 30 Aug 2023 • YuHang Zhou, Xuan Lu, Ge Gao, Qiaozhu Mei, Wei Ai
In this paper, we study how emoji usage influences developer participation and issue resolution in virtual workspaces.
1 code implementation • 3 Aug 2023 • YuHang Zhou, Jiangchao Yao, Feng Hong, Ya zhang, Yanfeng Wang
By dynamically manipulating the gradient during training based on these factors, BDR can effectively alleviate knowledge destruction and improve knowledge reconstruction.
no code implementations • 1 Jun 2023 • Jing Zhu, YuHang Zhou, Vassilis N. Ioannidis, Shengyi Qian, Wei Ai, Xiang Song, Danai Koutra
While Graph Neural Networks (GNNs) are remarkably successful in a variety of high-impact applications, we demonstrate that, in link prediction, the common practices of including the edges being predicted in the graph at training and/or test have outsized impact on the performance of low-degree nodes.
no code implementations • 25 May 2023 • Paiheng Xu, YuHang Zhou, Bang An, Wei Ai, Furong Huang
Given the growing concerns about fairness in machine learning and the impressive performance of Graph Neural Networks (GNNs) on graph data learning, algorithmic fairness in GNNs has attracted significant attention.
no code implementations • 18 Feb 2023 • YuHang Zhou, Suraj Maharjan, Beiye Liu
In this paper, we propose two methods to automatically design multiple prompts and integrate automatic verbalizer in SSL settings without sacrificing performance.
1 code implementation • 28 Dec 2022 • Zi'an Xu, Yin Dai, Fayu Liu, Weibing Chen, Yue Liu, Lifu Shi, Sheng Liu, YuHang Zhou
The development of deep learning models in medical image analysis is majorly limited by the lack of large-sized and well-annotated datasets.
1 code implementation • 16 Feb 2022 • Oana Ignat, Santiago Castro, YuHang Zhou, Jiajun Bao, Dandan Shan, Rada Mihalcea
We consider the task of temporal human action localization in lifestyle vlogs.
no code implementations • CVPR 2022 • Yuchen Li, Zixuan Li, Siyu Teng, Yu Zhang, YuHang Zhou, Yuchang Zhu, Dongpu Cao, Bin Tian, Yunfeng Ai, Zhe XuanYuan, Long Chen
The main contributions of the AutoMine dataset are as follows: 1. The first autonomous driving dataset for perception and localization in mine scenarios.
no code implementations • 30 Dec 2021 • Chenlin Shen, Guangda Huzhang, YuHang Zhou, Chen Liang, Qing Da
Our algorithm can straightforwardly optimize the linear programming in the prime space, and its solution can be simply applied by a stochastic strategy to fulfill the optimized objective and the constraints in expectation.
no code implementations • 5 Aug 2021 • Shixiang Feng, YuHang Zhou, Xiaoman Zhang, Ya zhang, Yanfeng Wang
A novel Multi-teacher Single-student Knowledge Distillation (MS-KD) framework is proposed, where the teacher models are pre-trained single-organ segmentation networks, and the student model is a multi-organ segmentation network.
no code implementations • 4 Aug 2021 • Liyuan Zhang, YuHang Zhou, Lei Zhang
State-of-the-art deep neural networks (DNNs) have been proved to have excellent performance on unsupervised domain adaption (UDA).
no code implementations • 9 Mar 2021 • YuHang Zhou, Xiaoman Zhang, Shixiang Feng, Ya zhang, Yanfeng
Specifically, given a pretrained $K$ organ segmentation model and a new single-organ dataset, we train a unified $K+1$ organ segmentation model without accessing any data belonging to the previous training stages.
no code implementations • 21 Oct 2020 • Yifan Hu, YuHang Zhou, Jun Xiao, Chao Wu
Federated learning(FL) is a rapidly growing field and many centralized and decentralized FL frameworks have been proposed.
no code implementations • 13 Oct 2020 • Xiaoman Zhang, Shixiang Feng, YuHang Zhou, Ya zhang, Yanfeng Wang
We demonstrate the effectiveness of our methods on two downstream tasks: i) Brain tumor segmentation, ii) Pancreas tumor segmentation.