Reproduce “InsTALL: Context-aware Instructional Task Assistance with Multi-modal Large Language Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Pha Nguyen, Sailik Sengupta, Girik Malik, Arshit Gupta, Bonan Min
This paper introduces InsTALL, a context-aware instructional task assistant that combines multi-modal large language models with procedural graphs for better understanding and assistance in multi-step tasks
Key insights
- Incorporating procedural graphs reduces reasoning burden on LLMs
- Graph-based approach enables better modeling of short-term dependencies between actions
- Combining multiple modalities (visual, text, graph) improves performance across all tasks
- Online graph construction during inference helps provide contextual assistance
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.