no code implementations • 19 Mar 2024 • Jiuhai Chen, Jonas Mueller
We introduce an automated data curation pipeline CLEAR (Confidence-based LLM Evaluation And Rectification) for instruction tuning datasets, that can be used with any LLM and fine-tuning procedure.
1 code implementation • 16 Feb 2024 • Ming Li, Jiuhai Chen, Lichang Chen, Tianyi Zhou
To examine DEBATunE, we curate the largest dataset of debate topics so far, which covers 710 controversial topics and corresponding arguments for each topic.
2 code implementations • 15 Feb 2024 • Ming Li, Lichang Chen, Jiuhai Chen, Shwai He, Jiuxiang Gu, Tianyi Zhou
Instruction tuning is critical to large language models (LLMs) for achieving better instruction following and task adaptation capabilities but its success heavily relies on the training data quality.
no code implementations • 11 Feb 2024 • Lichang Chen, Chen Zhu, Davit Soselia, Jiuhai Chen, Tianyi Zhou, Tom Goldstein, Heng Huang, Mohammad Shoeybi, Bryan Catanzaro
In this work, we study the issue of reward hacking on the response length, a challenge emerging in Reinforcement Learning from Human Feedback (RLHF) on LLMs.
2 code implementations • 18 Oct 2023 • Ming Li, Lichang Chen, Jiuhai Chen, Shwai He, Heng Huang, Jiuxiang Gu, Tianyi Zhou
Recent advancements in Large Language Models (LLMs) have expanded the horizons of natural language understanding and generation.
no code implementations • 30 Aug 2023 • Jiuhai Chen, Jonas Mueller
We introduce BSDetector, a method for detecting bad and speculative answers from a pretrained Large Language Model by estimating a numeric confidence score for any output it generated.
2 code implementations • 23 Aug 2023 • Ming Li, Yong Zhang, Zhitao Li, Jiuhai Chen, Lichang Chen, Ning Cheng, Jianzong Wang, Tianyi Zhou, Jing Xiao
In the realm of Large Language Models (LLMs), the balance between instruction data quality and quantity is a focal point.
1 code implementation • 5 Jun 2023 • Lichang Chen, Jiuhai Chen, Tom Goldstein, Heng Huang, Tianyi Zhou
Large language models~(LLMs) are instruction followers, but it can be challenging to find the best instruction for different situations, especially for black-box LLMs on which backpropagation is forbidden.
no code implementations • 6 Apr 2023 • Jiuhai Chen, Lichang Chen, Heng Huang, Tianyi Zhou
However, it is not clear whether CoT is still effective on more recent instruction finetuned (IFT) LLMs such as ChatGPT.
no code implementations • 14 Mar 2023 • Jiuhai Chen, Lichang Chen, Chen Zhu, Tianyi Zhou
Moreover, ICL (with and w/o CoT) using only one correct demo significantly outperforms all-demo ICL adopted by most previous works, indicating the weakness of LLMs in finding correct demo(s) for input queries, which is difficult to evaluate on the biased datasets.
no code implementations • 16 Jun 2022 • Jiuhai Chen, Jonas Mueller, Vassilis N. Ioannidis, Tom Goldstein, David Wipf
Graph Neural Networks (GNNs) with numerical node features and graph structure as inputs have demonstrated superior performance on various supervised learning tasks with graph data.
1 code implementation • 26 Oct 2021 • Jiuhai Chen, Jonas Mueller, Vassilis N. Ioannidis, Soji Adeshina, Yangkun Wang, Tom Goldstein, David Wipf
For supervised learning with tabular data, decision tree ensembles produced via boosting techniques generally dominate real-world applications involving iid training/test sets.
no code implementations • 26 Oct 2021 • Jiuhai Chen, Chen Zhu, Bin Dai
In this paper, we study how SSL can enhance the performance of the out-of-distribution (OOD) detection task.
Out-of-Distribution Detection Out of Distribution (OOD) Detection +1
no code implementations • ICLR 2022 • Yangkun Wang, Jiarui Jin, Weinan Zhang, Yongyi Yang, Jiuhai Chen, Quan Gan, Yong Yu, Zheng Zhang, Zengfeng Huang, David Wipf
In this regard, it has recently been proposed to use a randomly-selected portion of the training labels as GNN inputs, concatenated with the original node features for making predictions on the remaining labels.
no code implementations • 29 Sep 2021 • Mucong Ding, Kezhi Kong, Jiuhai Chen, John Kirchenbauer, Micah Goldblum, David Wipf, Furong Huang, Tom Goldstein
We observe that in most cases, we need both a suitable domain generalization algorithm and a strong GNN backbone model to optimize out-of-distribution test performance.
no code implementations • ICLR 2022 • Jiuhai Chen, Jonas Mueller, Vassilis N. Ioannidis, Soji Adeshina, Yangkun Wang, Tom Goldstein, David Wipf
Many practical modeling tasks require making predictions using tabular data composed of heterogeneous feature types (e. g., text-based, categorical, continuous, etc.).
1 code implementation • 14 Apr 2020 • Yiwei Wang, Jiuhai Chen, Chun Liu, Lulu Kang
Using the EVI framework, we can derive many existing Particle-based Variational Inference (ParVI) methods, including the popular Stein Variational Gradient Descent (SVGD) approach.