Contextcite: Attributing model generation to context

B Cohen-Wang, H Shah, K Georgiev… - arXiv preprint arXiv …, 2024 - arxiv.org
How do language models use information provided as context when generating a
response? Can we infer whether a particular generated statement is actually grounded in …

A primer on the inner workings of transformer-based language models

J Ferrando, G Sarti, A Bisazza… - arXiv preprint arXiv …, 2024 - arxiv.org
The rapid progress of research aimed at interpreting the inner workings of advanced
language models has highlighted a need for contextualizing the insights gained from years …

Unifying corroborative and contributive attributions in large language models

T Worledge, JH Shen, N Meister… - … IEEE Conference on …, 2024 - ieeexplore.ieee.org
As businesses, products, and services spring up around large language models, the
trustworthiness of these models hinges on the verifiability of their outputs. However, methods …

Editable concept bottleneck models

L Hu, C Ren, Z Hu, H Lin, CL Wang, H Xiong… - arXiv preprint arXiv …, 2024 - arxiv.org
Concept Bottleneck Models (CBMs) have garnered much attention for their ability to
elucidate the prediction process through a human-understandable concept layer. However …

Towards User-Focused Research in Training Data Attribution for Human-Centered Explainable AI

E Nguyen, J Bertram, E Kortukov, JY Song… - arXiv preprint arXiv …, 2024 - arxiv.org
While Explainable AI (XAI) aims to make AI understandable and useful to humans, it has
been criticised for relying too much on formalism and solutionism, focusing more on …

Advances and open challenges in federated learning with foundation models

C Ren, H Yu, H Peng, X Tang, A Li, Y Gao… - arXiv preprint arXiv …, 2024 - arxiv.org
The integration of Foundation Models (FMs) with Federated Learning (FL) presents a
transformative paradigm in Artificial Intelligence (AI), offering enhanced capabilities while …

Reciprocal learning

J Rodemann, C Jansen, G Schollmeyer - arXiv preprint arXiv:2408.06257, 2024 - arxiv.org
We demonstrate that a wide array of machine learning algorithms are specific instances of
one single paradigm: reciprocal learning. These instances range from active learning over …

What is Your Data Worth to GPT? LLM-Scale Data Valuation with Influence Functions

SK Choe, H Ahn, J Bae, K Zhao, M Kang… - arXiv preprint arXiv …, 2024 - arxiv.org
Large language models (LLMs) are trained on a vast amount of human-written data, but data
providers often remain uncredited. In response to this issue, data valuation (or data …

Enhancing Data Quality in Federated Fine-Tuning of Foundation Models

W Zhao, Y Du, ND Lane, S Chen, Y Wang - arXiv preprint arXiv …, 2024 - arxiv.org
In the current landscape of foundation model training, there is a significant reliance on public
domain data, which is nearing exhaustion according to recent research. To further scale up …

Do Influence Functions Work on Large Language Models?

Z Li, W Zhao, Y Li, J Sun - arXiv preprint arXiv:2409.19998, 2024 - arxiv.org
Influence functions aim to quantify the impact of individual training data points on a model's
predictions. While extensive research has been conducted on influence functions in …