Reproduce “Multi-turn Conversational AI from Text to Multimodal Interaction: Data, Models, Evaluation, and Open Challenges” 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
Syeda Faiza Ahmed Sara, Zien Sheikh Ali, Hunzalah Hassan Bhatti, Firoj Alam, Shammur Absar Chowdhury
The authors survey session-level multi-turn conversational AI across text, speech, vision, video, tools, and evaluation, using full sessions as a frame. They report 43 of 52 dataset entries are English-only outside cultural and cross-lingual resources, with 18 text-image, 8 spoken, and 6 video-oriented entries. Four of five cultural-block entries are single-turn. These counts describe uneven coverage, not capability or causation.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.