4 papers
Whose Name Comes Up? Auditing LLM-Based Scholar Recommendations
Daniele Barolo, Chiara Valentin, Fariba Karimi +3
This paper evaluates the performance of six open-weight LLMs (llama3-8b, llama3.1-8b, gemma2-9b, mixtral-8x7b, llama3-70b, llama3.1-70b) in recommending experts in physics across f…
REFINE-LM: Mitigating Language Model Stereotypes via Reinforcement Learning
Rameez Qureshi, Naïm Es-Sebbani, Luis Galárraga +3
With the introduction of (large) language models, there has been significant concern about the unintended bias such models may inherit from their training data. A number of studies…
Shaping Up SHAP: Enhancing Stability through Layer-Wise Neighbor Selection
Gwladys Kelodjou, Laurence Rozé, Véronique Masson +4
Machine learning techniques, such as deep learning and ensemble methods, are widely used in various domains due to their ability to handle complex real-world tasks. However, their…
Does It Make Sense to Explain a Black Box With Another Black Box?
Julien Delaunay, Luis Galárraga, Christine Largouët
Although counterfactual explanations are a popular approach to explain ML black-box classifiers, they are less widespread in NLP. Most methods find those explanations by iterativel…