Overview
The third workshop on Visually Grounded Interaction and Language (ViGIL) is set to take place at NeurIPS 2019 in Vancouver, Canada. This event aims to explore the intersection of language, vision, and interaction, drawing insights from various disciplines including machine learning, cognitive science, and neuroscience. The workshop provides a platform for researchers to discuss innovative approaches to grounding language in multimodal contexts.
Background & Relevance
The field of natural language understanding has traditionally relied on statistical models trained on text-only data. This method, while effective, often leads to a limited understanding of semantics, as it does not incorporate the rich, multimodal environments in which human communication occurs. The ViGIL workshop seeks to address this gap by fostering discussions on how to integrate visual information with language processing, thereby enhancing the semantic grounding of language models. This is particularly relevant as advancements in deep learning and reinforcement learning continue to reshape the landscape of AI research.
Key Details
- Event: ViGIL 2019 Workshop on Visually Grounded Interaction and Language
- Dates: December 13-14, 2019
- Location: Vancouver, Canada
- Submission Deadline: September 15, 2019, Midnight Pacific Time
- Final Decisions: September 22, 2019
- Website: ViGIL Workshop
- Contact: vigilworkshop@gmail.com
Eligibility & Participation
The workshop invites contributions from researchers across multiple fields, including machine learning, computer vision, natural language processing, and cognitive science. It targets those interested in the challenges and methodologies related to visually grounded language understanding and interaction.
Submission or Application Guidelines
Submissions should be up to four pages in length, excluding references and supplementary material, and must adhere to the NeurIPS formatting guidelines. The review process will be double-blind to ensure impartiality. Accepted papers will be presented in joint poster sessions, with outstanding submissions selected for spotlight presentations. Papers previously published in non-ML conferences are also welcome, provided they align with the workshop’s scope.
To submit, please visit: CMT Submission. For inquiries, contact vigilworkshop@gmail.com.
Additional Context / Real-World Relevance
The integration of language and vision is crucial for developing AI systems that can understand and interact with the world in a human-like manner. This workshop emphasizes the importance of grounding language in perceptual experiences, which is essential for advancing AI capabilities in real-world applications such as robotics and human-computer interaction. By bridging insights from cognitive science and machine learning, the workshop aims to inspire innovative research directions.
Conclusion
The ViGIL 2019 workshop presents a unique opportunity for researchers to engage with pressing issues at the intersection of language and vision. Participants are encouraged to submit their work and contribute to the growing dialogue on how to effectively ground language in multimodal contexts. This is a chance to share ideas, collaborate, and push the boundaries of what is possible in AI research.
Category: CFP & Deadlines
Tags: visually grounded interaction, natural language processing, machine learning, computer vision, cognitive science, multimodal ai, deep learning, reinforcement learning, language acquisition
