no code implementations • 2 Jan 2025 • Yixu Wang, Tianle Gu, Yan Teng, Yingchun Wang, Xingjun Ma
In this work, we introduce a new defense paradigm called attack as defense which modifies the model's output to be poisonous such that any malicious users that attempt to use the output to train a substitute model will be poisoned.
no code implementations • 8 Dec 2024 • Chao Yang, Chaochao Lu, Yingchun Wang, BoWen Zhou
In this position paper, we propose the \textit{AI-\textbf{$45^{\circ}$} Law} as a guiding principle for a balanced roadmap toward trustworthy AGI, and introduce the \textit{Causal Ladder of Trustworthy AGI} as a practical framework.
1 code implementation • 21 Oct 2024 • Lingyu Li, Yixu Wang, Haiquan Zhao, Shuqi Kong, Yan Teng, Chunbo Li, Yingchun Wang
The ability to adapt beliefs or behaviors in response to unexpected outcomes, reflection, is fundamental to intelligent systems' interaction with the world.
1 code implementation • 18 Sep 2024 • Tianle Gu, Kexin Huang, Ruilin Luo, Yuanqi Yao, Yujiu Yang, Yan Teng, Yingchun Wang
LLM Unlearning, a post-hoc approach to remove this information from trained LLMs, offers a promising solution to mitigate these risks.
1 code implementation • 24 Jul 2024 • Junyu Li, Ye Zhang, Wen Shu, Xiaobing Feng, Yingchun Wang, Pengju Yan, Xiaolin Li, Chulin Sha, Min He
Multiple instance learning (MIL) has been successfully applied for whole slide images (WSIs) analysis in computational pathology, enabling a wide range of prediction tasks from tumor subtyping to inferring genetic mutations and multi-omics biomarkers.
no code implementations • 15 Jul 2024 • Xuhong Wang, Haoyu Jiang, Yi Yu, Jingru Yu, Yilun Lin, Ping Yi, Yingchun Wang, Yu Qiao, Li Li, Fei-Yue Wang
Large Language Models (LLMs) are increasingly integrated into diverse industries, posing substantial security risks due to unauthorized replication and misuse.
3 code implementations • 21 Jun 2024 • Haiquan Zhao, Lingyu Li, Shisong Chen, Shuqi Kong, Jiaan Wang, Kexin Huang, Tianle Gu, Yixu Wang, Wang Jian, Dandan Liang, Zhixu Li, Yan Teng, Yanghua Xiao, Yingchun Wang
Inspired by the awesome development of role-playing agents, we propose an ESC Evaluation framework (ESC-Eval), which uses a role-playing agent to interact with ESC models, followed by a manual evaluation of the interactive dialogues.
1 code implementation • 11 Jun 2024 • Tianle Gu, Zeyang Zhou, Kexin Huang, Dandan Liang, Yixu Wang, Haiquan Zhao, Yuanqi Yao, Xingge Qiao, Keqing Wang, Yujiu Yang, Yan Teng, Yu Qiao, Yingchun Wang
In this paper, we present MLLMGuard, a multidimensional safety evaluation suite for MLLMs, including a bilingual image-text evaluation dataset, inference utilities, and a lightweight evaluator.
no code implementations • 26 Jan 2024 • Chaochao Lu, Chen Qian, Guodong Zheng, Hongxing Fan, Hongzhi Gao, Jie Zhang, Jing Shao, Jingyi Deng, Jinlan Fu, Kexin Huang, Kunchang Li, Lijun Li, LiMin Wang, Lu Sheng, Meiqi Chen, Ming Zhang, Qibing Ren, Sirui Chen, Tao Gui, Wanli Ouyang, Yali Wang, Yan Teng, Yaru Wang, Yi Wang, Yinan He, Yingchun Wang, Yixu Wang, Yongting Zhang, Yu Qiao, Yujiong Shen, Yurong Mou, Yuxi Chen, Zaibin Zhang, Zhelun Shi, Zhenfei Yin, Zhipin Wang
Multi-modal Large Language Models (MLLMs) have shown impressive abilities in generating reasonable responses with respect to multi-modal contents.
1 code implementation • 12 Nov 2023 • Kexin Huang, Xiangyang Liu, Qianyu Guo, Tianxiang Sun, Jiawei Sun, Yaru Wang, Zeyang Zhou, Yixu Wang, Yan Teng, Xipeng Qiu, Yingchun Wang, Dahua Lin
The widespread adoption of large language models (LLMs) across various regions underscores the urgent need to evaluate their alignment with human values.
1 code implementation • 10 Nov 2023 • Yixu Wang, Yan Teng, Kexin Huang, Chengqi Lyu, Songyang Zhang, Wenwei Zhang, Xingjun Ma, Yu-Gang Jiang, Yu Qiao, Yingchun Wang
The growing awareness of safety concerns in large language models (LLMs) has sparked considerable interest in the evaluation of safety.
no code implementations • 19 May 2023 • Yingchun Wang, Jingcai Guo, Yi Liu, Song Guo, Weizhan Zhang, Xiangyong Cao, Qinghua Zheng
Based on the idea that in-distribution (ID) data with spurious features may have a lower experience risk, in this paper, we propose a novel Spurious Feature-targeted model Pruning framework, dubbed SFP, to automatically explore invariant substructures without referring to the above drawbacks.
no code implementations • 9 Feb 2023 • Yingchun Wang, Jingcai Guo, Song Guo, Weizhan Zhang
Mixed-precision quantization mostly predetermines the model bit-width settings before actual training due to the non-differential bit-width sampling process, obtaining sub-optimal performance.
no code implementations • 9 Feb 2023 • Yingchun Wang, Jingcai Guo, Jie Zhang, Song Guo, Weizhan Zhang, Qinghua Zheng
Federated learning (FL) is an emerging technique that trains massive and geographically distributed edge data while maintaining privacy.
no code implementations • 19 Dec 2022 • Yingchun Wang, Jingcai Guo, Song Guo, Weizhan Zhang, Jie Zhang
Recent studies show that even highly biased dense networks contain an unbiased substructure that can achieve better out-of-distribution (OOD) generalization than the original model.
no code implementations • 7 Dec 2022 • Yingchun Wang, Song Guo, Jingcai Guo, Weizhan Zhang, Yida Xu, Jie Zhang, Yi Liu
Extensive experiments based on small Cifar-10 and large-scaled ImageNet demonstrate that our method can obtain sparser networks with great generalization performance while providing quantified reliability for the pruned model.
no code implementations • 14 Nov 2022 • Yi Liu, Song Guo, Jie Zhang, Qihua Zhou, Yingchun Wang, Xiaohan Zhao
We prove that FedFoA is a model-agnostic training framework and can be easily compatible with state-of-the-art unsupervised FL methods.