no code implementations • 24 May 2024 • Jingcheng Deng, Zihao Wei, Liang Pang, Hanxing Ding, HuaWei Shen, Xueqi Cheng
Firstly, in the layer dimension, we discard the "knowledge locating" step and treat first few layers as the key, which expand knowledge storage through layers to break the "knowledge stored locally" assumption.
no code implementations • 26 Apr 2024 • Shuchang Tao, Liuyi Yao, Hanxing Ding, Yuexiang Xie, Qi Cao, Fei Sun, Jinyang Gao, HuaWei Shen, Bolin Ding
Specifically, the order-preserving reward incentivizes the model to verbalize greater confidence for responses of higher quality to align the order of confidence and quality.
2 code implementations • 7 Apr 2024 • Zihao Wei, Jingcheng Deng, Liang Pang, Hanxing Ding, HuaWei Shen, Xueqi Cheng
We evaluate the multilingual knowledge editing generalization capabilities of existing methods on MLaKE.
1 code implementation • 20 Feb 2024 • Zihao Wei, Liang Pang, Hanxing Ding, Jingcheng Deng, HuaWei Shen, Xueqi Cheng
The premise of localization results in an incomplete knowledge editing, whereas an isolated assumption may impair both other knowledge and general abilities.
no code implementations • 16 Feb 2024 • Hanxing Ding, Liang Pang, Zihao Wei, HuaWei Shen, Xueqi Cheng
A careful and balanced integration of the parametric knowledge within LLMs with external information is crucial to alleviate hallucinations.
1 code implementation • 22 May 2023 • Hanxing Ding, Liang Pang, Zihao Wei, HuaWei Shen, Xueqi Cheng, Tat-Seng Chua
Multi-aspect controllable text generation aims to generate fluent sentences that possess multiple desired attributes simultaneously.