Paper Title: A Probabilistic Hard Concept Bottleneck for Steerable Generative Models
Authors: María Martínez-García, Ricardo Vazquez Alvarez, Alejandro Lancho, Pablo M. Olmos, Isabel Valera
Link: https://openreview.net/pdf/d3f49948fcc28f692fbc2c3bf36afe5a0ecc7853.pdf
Focus: This paper introduces a new framework that embeds human-understandable concepts into generative models. It allows users to control and steer image generation precisely by altering specific visual attributes.
And
Paper Title: Bridging Fairness and Explainability: Can Input-Based Explanations Promote Fairness in Hate Speech Detection?
Authors: Yifan Wang, Mayank Jobanputra, Ji-Ung Lee, Soyoung Oh, Isabel Valera, Vera Demberg
Link: https://openreview.net/forum?id=fPMu3Afv3s
Focus: This work investigates whether showing why a model flagged text as hate speech reduces algorithmic bias. It evaluates how feature explanations affect fairness across different demographic groups.