awesome-multimodal-ml

Why we included this project

Multimodal machine learning, where a model has to make sense of vision, language, audio, and other inputs at once, is a sprawling field, and this reading list is a good way to get a handle on it. It is organized around the problems researchers actually work on, such as representation learning, fusion, alignment, pretraining, and crossmodal retrieval, then branches into application areas like visual question answering, healthcare, robotics, and autonomous driving. Every entry links to the paper and, when it exists, the code, so you can go from reading to reproducing results without hunting for the source. The maintainers are CMU researchers, and the list pulls together survey papers, workshop materials, tutorials, and full course content, which makes it a solid starting point for a graduate student, a new team member, or a practitioner trying to get oriented before picking a direction. It is a curated index rather than runnable software, but for building a grounded picture of the field it does the job better than most.