Reproducibility study of "LICO: Explainable Models with Language-Image Consistency"
This is an incremental reproducibility study addressing the machine learning community's need for verification of interpretability research findings.
This paper attempted to reproduce the claims of LICO, a method for enhancing interpretability and image classification performance using language supervision, but found it did not consistently improve classification or interpretability metrics as originally reported.
The growing reproducibility crisis in machine learning has brought forward a need for careful examination of research findings. This paper investigates the claims made by Lei et al. (2023) regarding their proposed method, LICO, for enhancing post-hoc interpretability techniques and improving image classification performance. LICO leverages natural language supervision from a vision-language model to enrich feature representations and guide the learning process. We conduct a comprehensive reproducibility study, employing (Wide) ResNets and established interpretability methods like Grad-CAM and RISE. We were mostly unable to reproduce the authors' results. In particular, we did not find that LICO consistently led to improved classification performance or improvements in quantitative and qualitative measures of interpretability. Thus, our findings highlight the importance of rigorous evaluation and transparent reporting in interpretability research.