Efficient Multimodal Question Answering (EMM-QA)
Efficient multimodal question answering in the era of large language models.
EMM-QA is an ICML 2026 workshop focused on question answering systems that must balance accuracy, efficiency, and adaptability across multiple input modalities. The workshop brings together researchers from academia and industry working on knowledge-intensive multimodal systems that operate under practical resource constraints.
Rather than focusing only on larger models, the workshop emphasizes methods that make multimodal question answering usable in real settings, including retrieval-augmented systems, compact models, efficient inference, and human-in-the-loop evaluation.
Join the community on Discord.
Scope
The workshop is centered on efficient multimodal question answering. It also welcomes closely related work on multimodal retrieval, reasoning, evaluation, benchmarking, and efficient inference when those contributions are clearly connected to question answering or other knowledge-intensive multimodal tasks.
Workshop Format
The workshop is planned as a one-day event combining:
- Contributed papers
- Poster presentations
- Shared-task highlights
- A live human-computer question answering event
- A panel discussion
The workshop will also serve as the venue where we announce the winning systems from the QANTA 2026 computer competition.
Shared Task
The shared task focuses on efficient multimodal question answering over inputs such as text, images, tables, and audio. Participants will build systems that answer open-domain questions while balancing answer quality against practical constraints such as model footprint and inference cost.
The competition component extends earlier EfficientQA ideas to multimodal settings and includes both automatic evaluation and human-centered analysis of when people should trust, verify, or override system outputs.
Audience
EMM-QA is intended for researchers working on question answering, multimodal machine learning and NLP, retrieval, efficient modeling, and LLMs across both academia and industry.
Schedule
A detailed workshop schedule will be posted here soon.
Confirmed Keynote Speakers
- Sewon Min
- Mrinmaya Sachan
Organizers
- Jordan Boyd-Graber, University of Maryland
- Martin Fajcik, Brno University of Technology
- George Jojo Boateng, ETH Zurich / Kwame AI
- Ikuya Yamada, Studio Ousia / Tohoku University / Nagoya University / RIKEN
- Chen Zhao, NYU Shanghai