1 code implementation • 14 Mar 2024 • Fan Zhang, Wei Qin, Weijieying Ren, Lei Wang, Zetong Chen, Richang Hong
Additionally, We find that most of the solutions to long-tailed problems are still biased towards head classes in the end, and we propose a simple and post hoc prediction re-balancing strategy to further mitigate the basis toward head class.
1 code implementation • 29 Feb 2024 • Weijieying Ren, Xinlong Li, Lei Wang, Tianxiang Zhao, Wei Qin
Through extensive experiments, we uncover the mode connectivity phenomenon in the LLMs continual learning scenario and find that it can strike a balance between plasticity and stability.
no code implementations • 11 Jan 2024 • Weijieying Ren, Vasant G Honavar
A key challenge in the continual learning setting is to efficiently learn a sequence of tasks without forgetting how to perform previously learned tasks.
no code implementations • 5 Sep 2023 • Weijieying Ren, Tianxiang Zhao, Wei Qin, Kunpeng Liu
Discovering the shifted behaviors and the evolving patterns underlying the streaming data are important to understand the dynamic system.
no code implementations • 16 Aug 2023 • Qinghua Shen, Weijieying Ren, Wei Qin
Learning a single model could loss discriminative information for each graph task while the model expansion scheme suffers from high model complexity.
no code implementations • 9 May 2023 • Wei Qin, Zetong Chen, Lei Wang, Yunshi Lan, Weijieying Ren, Richang Hong
This paper proposes a new depression detection system based on LLMs that is both interpretable and interactive.
no code implementations • 31 Oct 2022 • Weijieying Ren, Lei Wang, Kunpeng Liu, Ruocheng Guo, Lim Ee Peng, Yanjie Fu
We present a gradient perspective to understand two negative impacts of popularity bias in recommendation model optimization: (i) the gradient direction of popular item embeddings is closer to that of positive interactions, and (ii) the magnitude of positive gradient for popular items are much greater than that of unpopular items.
no code implementations • 25 May 2022 • Weijieying Ren, Pengyang Wang, Xiaolin Li, Charles E. Hughes, Yanjie Fu
In many scenarios, 1) data streams are generated in real time; 2) labeled data are expensive and only limited labels are available in the beginning; 3) real-world data is not always i. i. d.