Jean-Luc Picard at Touché 2023: Comparing Image Generation, Stance Detection and Feature Matching for Image Retrieval for Arguments
arXiv:2307.09172v14 citationsh-index: 4
Originality Synthesis-oriented
AI Analysis
This work addresses image retrieval for arguments, but it is incremental as it shows no improvement over existing methods.
The paper tackled the task of image retrieval for arguments by testing pipelines combining image generation, stance detection, and feature matching, finding that their approaches performed similarly to a baseline without providing specific numerical gains.
Participating in the shared task "Image Retrieval for arguments", we used different pipelines for image retrieval containing Image Generation, Stance Detection, Preselection and Feature Matching. We submitted four different runs with different pipeline layout and compare them to given baseline. Our pipelines perform similarly to the baseline.