{"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/mamba-fusion-learning-actions-through","title":"Mamba Fusion: Learning Actions Through Questioning","arxiv_id":"2409.11513","date":"2024-09-17","proceeding":null,"authors":["Zhikang Dong","Apoorva Beedu","Jason Sheinkopf","Irfan Essa"],"abstract":"Video Language Models (VLMs) are crucial for generalizing across diverse tasks and using language cues to enhance learning. While transformer-based architectures have been the de facto in vision-language training, they face challenges like quadratic computational complexity, high GPU memory usage, and difficulty with long-term dependencies. To address these limitations, we introduce MambaVL, a novel model that leverages recent advancements in selective state space modality fusion to efficiently capture long-range dependencies and learn joint representations for vision and language data. MambaVL utilizes a shared state transition matrix across both modalities, allowing the model to capture information about actions from multiple perspectives within the scene. Furthermore, we propose a question-answering task that helps guide the model toward relevant cues. These questions provide critical information about actions, objects, and environmental context, leading to enhanced performance. As a result, MambaVL achieves state-of-the-art performance in action recognition on the Epic-Kitchens-100 dataset and outperforms baseline methods in action anticipation.","url_abs":"https://arxiv.org/abs/2409.11513v2","url_pdf":"https://arxiv.org/pdf/2409.11513v2.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":"mamba-fusion-learning-actions-through","repo_url":"https://github.com/dongzhikang/mambavl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"action-anticipation","task_name":"Action Anticipation"},{"task_slug":"action-recognition-in-videos","task_name":"Action Recognition"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"mamba","task_name":"Mamba"},{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}