1 code implementation • 11 Dec 2024 • Chung-En Sun, Tuomas Oikarinen, Berk Ustun, Tsui-Wei Weng
We investigate two essential tasks in the NLP domain: text classification and text generation.
1 code implementation • 9 Nov 2024 • Aya Abdelsalam Ismail, Tuomas Oikarinen, Amy Wang, Julius Adebayo, Samuel Stanton, Taylor Joren, Joseph Kleinhenz, Allen Goodman, Héctor Corrada Bravo, Kyunghyun Cho, Nathan C. Frey
We introduce Concept Bottleneck Protein Language Models (CB-pLM), a generative masked language model with a layer where each neuron corresponds to an interpretable concept.
1 code implementation • 5 Jul 2024 • Chung-En Sun, Tuomas Oikarinen, Tsui-Wei Weng
We introduce the Concept Bottleneck Large Language Model (CB-LLM), a pioneering approach to creating inherently interpretable Large Language Models (LLMs).
1 code implementation • 10 May 2024 • Tuomas Oikarinen, Tsui-Wei Weng
In recent years many methods have been developed to understand the internal workings of neural networks, often by describing the function of individual neurons in the model.
1 code implementation • 20 Mar 2024 • Nicholas Bai, Rahul A. Iyer, Tuomas Oikarinen, Tsui-Wei Weng
In this paper, we propose Describe-and-Dissect (DnD), a novel method to describe the roles of hidden neurons in vision networks.
1 code implementation • ICCV 2023 • Divyansh Srivastava, Tuomas Oikarinen, Tsui-Wei Weng
The inability of DNNs to explain their black-box behavior has led to a recent surge of explainability methods.
no code implementations • 9 Oct 2023 • Justin Lee, Tuomas Oikarinen, Arjun Chatha, Keng-Chi Chang, Yilan Chen, Tsui-Wei Weng
Recent advances have greatly increased the capabilities of large language models (LLMs), but our understanding of the models and their safety has not progressed as fast.
no code implementations • 26 Apr 2023 • Mohammad Ali Khan, Tuomas Oikarinen, Tsui-Wei Weng
In this work, we propose a general framework called Concept-Monitor to help demystify the black-box DNN training processes automatically using a novel unified embedding space and concept diversity metric.
2 code implementations • 12 Apr 2023 • Tuomas Oikarinen, Subhro Das, Lam M. Nguyen, Tsui-Wei Weng
Motivated by these challenges, we propose Label-free CBM which is a novel framework to transform any neural network into an interpretable CBM without labeled concept data, while retaining a high accuracy.
2 code implementations • 23 Apr 2022 • Tuomas Oikarinen, Tsui-Wei Weng
Finally CLIP-Dissect is computationally efficient and can label all neurons from five layers of ResNet-50 in just 4 minutes, which is more than 10 times faster than existing methods.
2 code implementations • NeurIPS 2021 • Tuomas Oikarinen, Wang Zhang, Alexandre Megretski, Luca Daniel, Tsui-Wei Weng
To address this issue, we propose RADIAL-RL, a principled framework to train reinforcement learning agents with improved robustness against $l_p$-norm bounded adversarial attacks.