Explanipedia

Dual-Balancing for Multi-Task Learning Open

Baijiong Lin, Weisen Jiang, Pengguang Chen, Feiyang Ye, Yu Zhang , et al. · 2025

MTMamba++: Enhancing Multi-Task Dense Scene Understanding via Mamba-Based Decoders Open

Baijiong Lin, Weisen Jiang, Pengguang Chen, Shu Liu, Ying-Cong Chen · 2024

Multi-task dense scene understanding, which trains a model for multiple dense prediction tasks, has a wide range of application scenarios. Capturing long-range dependency and enhancing cross-task interactions are crucial to multi-task dens…

MTMamba: Enhancing Multi-Task Dense Scene Understanding by Mamba-Based Decoders Open

Baijiong Lin, Weisen Jiang, Pengguang Chen, Yu Zhang, Shu Liu , et al. · 2024

Multi-task dense scene understanding, which learns a model for multiple dense prediction tasks, has a wide range of application scenarios. Modeling long-range dependency and enhancing cross-task interactions are crucial to multi-task dense…

MR-Ben: A Meta-Reasoning Benchmark for Evaluating System-2 Thinking in LLMs Open

Zhongshen Zeng, Yinhong Liu, Yingjia Wan, Jingyao Li, Pengguang Chen , et al. · 2024

Large language models (LLMs) have shown increasing capability in problem-solving and decision-making, largely based on the step-by-step chain-of-thought reasoning processes. However, evaluating these reasoning abilities has become increasi…

VLPose: Bridging the Domain Gap in Pose Estimation with Language-Vision Tuning Open

Jingyao Li, Pengguang Chen, Xuan Ju, Hong Xu, Jiaya Jia · 2024

Thanks to advances in deep learning techniques, Human Pose Estimation (HPE) has achieved significant progress in natural scenarios. However, these models perform poorly in artificial scenarios such as painting and sculpture due to the doma…

MOODv2: Masked Image Modeling for Out-of-Distribution Detection Open

Jingyao Li, Pengguang Chen, Shaozuo Yu, Shu Liu, Jiaya Jia · 2024

The crux of effective out-of-distribution (OOD) detection lies in acquiring a robust in-distribution (ID) representation, distinct from OOD samples. While previous methods predominantly leaned on recognition-based techniques for this purpo…

MR-GSM8K: A Meta-Reasoning Benchmark for Large Language Model Evaluation Open

Zhongshen Zeng, Pengguang Chen, Haiyun Jiang, Jiaya Jia · 2023

In this work, we introduce a novel evaluation paradigm for Large Language Models (LLMs) that compels them to transition from a traditional question-answering role, akin to a student, to a solution-scoring role, akin to a teacher. This para…

BAL: Balancing Diversity and Novelty for Active Learning Open

Jingyao Li, Pengguang Chen, Shaozuo Yu, Shu Liu, Jiaya Jia · 2023

The objective of Active Learning is to strategically label a subset of the dataset to maximize performance within a predetermined labeling budget. In this study, we harness features acquired through self-supervised learning. We introduce a…

MoTCoder: Elevating Large Language Models with Modular of Thought for Challenging Programming Tasks Open

Jingyao Li, Pengguang Chen, Jiaya Jia · 2023

Large Language Models (LLMs) have showcased impressive capabilities in handling straightforward programming tasks. However, their performance tends to falter when confronted with more challenging programming problems. We observe that conve…

Defect Spectrum: A Granular Look of Large-Scale Defect Datasets with Rich Semantics Open

Shuai Yang, Zhifei Chen, Pengguang Chen, Xi Fang, Shuming Liu , et al. · 2023

Defect inspection is paramount within the closed-loop manufacturing system. However, existing datasets for defect inspection often lack precision and semantic granularity required for practical applications. In this paper, we introduce the…

Dual-Balancing for Multi-Task Learning Open

Baijiong Lin, Weisen Jiang, Feiyang Ye, Yu Zhang, Pengguang Chen , et al. · 2023

Multi-task learning aims to learn multiple related tasks simultaneously and has achieved great success in various fields. However, the disparity in loss and gradient scales among tasks often leads to performance compromises, and the balanc…

TagCLIP: Improving Discrimination Ability of Open-Vocabulary Semantic Segmentation Open

Jingyao Li, Pengguang Chen, Shengju Qian, Jiaya Jia · 2023

Contrastive Language-Image Pre-training (CLIP) has recently shown great promise in pixel-level zero-shot learning tasks. However, existing approaches utilizing CLIP's text and patch embeddings to generate semantic masks often misidentify i…

Rethinking Out-of-distribution (OOD) Detection: Masked Image Modeling is All You Need Open

Jingyao Li, Pengguang Chen, Shaozuo Yu, Zexin He, Shu Liu , et al. · 2023

The core of out-of-distribution (OOD) detection is to learn the in-distribution (ID) representation, which is distinguishable from OOD samples. Previous work applied recognition-based methods to learn the ID features, which tend to learn s…

Adversarial Attacks on ML Defense Models Competition Open

Yinpeng Dong, Qian Fu, Xiao Yang, Wenzhao Xiang, Tianyu Pang , et al. · 2021

Due to the vulnerability of deep neural networks (DNNs) to adversarial examples, a large number of defense techniques have been proposed to alleviate this problem in recent years. However, the progress of building more robust models is usu…

Deep Structured Instance Graph for Distilling Object Detectors Open

Yixin Chen, Pengguang Chen, Shu Liu, Liwei Wang, Jiaya Jia · 2021

Effectively structuring deep knowledge plays a pivotal role in transfer from teacher to student, especially in semantic vision tasks. In this paper, we present a simple knowledge structure to exploit and encode information inside the detec…

Exploring and Improving Mobile Level Vision Transformers Open

Pengguang Chen, Yixin Chen, Shu Liu, Ming-Chang Yang, Jiaya Jia · 2021

We study the vision transformer structure in the mobile level in this paper, and find a dramatic performance drop. We analyze the reason behind this phenomenon, and propose a novel irregular patch embedding module and adaptive patch fusion…

Distilling Knowledge via Knowledge Review Open

Pengguang Chen, Shu Liu, Hengshuang Zhao, Jiaya Jia · 2021

Knowledge distillation transfers knowledge from the teacher network to the student one, with the goal of greatly improving the performance of the student network. Previous methods mostly focus on proposing feature transformation and loss f…

Jigsaw Clustering for Unsupervised Visual Representation Learning Open

Pengguang Chen, Shu Liu, Jiaya Jia · 2021

Unsupervised representation learning with contrastive learning achieved great success. This line of methods duplicate each training batch to construct contrastive pairs, making each training batch and its augmented version forwarded simult…

Pengguang Chen YOU? Author Swipe