{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/explaining-multi-modal-large-language-models","title":"Explaining Multi-modal Large Language Models by Analyzing their Vision Perception","arxiv_id":"2405.14612","date":"2024-05-23","proceeding":null,"authors":["Loris Giulivi","Giacomo Boracchi"],"abstract":"Multi-modal Large Language Models (MLLMs) have demonstrated remarkable capabilities in understanding and generating content across various modalities, such as images and text. However, their interpretability remains a challenge, hindering their adoption in critical applications. This research proposes a novel approach to enhance the interpretability of MLLMs by focusing on the image embedding component. We combine an open-world localization model with a MLLM, thus creating a new architecture able to simultaneously produce text and object localization outputs from the same vision embedding. The proposed architecture greatly promotes interpretability, enabling us to design a novel saliency map to explain any output token, to identify model hallucinations, and to assess model biases through semantic adversarial perturbations.","url_abs":"https://arxiv.org/abs/2405.14612v2","url_pdf":"https://arxiv.org/pdf/2405.14612v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"explaining-multi-modal-large-language-models","repo_url":"https://github.com/loris2222/ExplainingMLLMs","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"jax","reach":null}],"tasks":[{"task_slug":"object-localization","task_name":"Object Localization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}