Search papers, labs, and topics across Lattice.
This paper introduces Reflective Dialogue (RD), a novel inference-time adaptation method for Video Question Answering that uses multi-turn conversations between Teacher and Solver agents. The Teacher agent poses questions from a support set and provides correctness feedback, while the Solver agent answers and offers visual grounding explanations. By using the dialogue history as context, the method achieves state-of-the-art performance on the EgoCross benchmark without fine-tuning.
Forget fine-tuning: this method achieves impressive VQA adaptation simply by having two agents talk it out.
Various approaches have been proposed to adapt Vision-Language Models (VLMs) to specialized domains for Video Question Answering, including fine-tuning and in-context learning. However, acquiring task-specific knowledge at the inference phase from only a small labeled support set without fine-tuning remains a challenge. In this paper, we propose a method that achieves adaptation solely through inference-time context injection. Our method first constructs a Reflective Dialogue (RD) -- a multi-turn conversation between two agents, in which Teacher poses each support question and delivers correctness feedback, and Solver answers and provides visual grounding explanations (or reflections) for both correct and incorrect answers. This dialogue history is then used as context at the inference phase. Experiments on the EgoCross benchmark demonstrate that our method outperforms both a baseline zero-shot setting and a standard in-context learning approach that passes support set examples directly, achieving 3rd place in the Open-source Track of the 1st Cross-Domain EgoCross Challenge at the CVPR 2026 EgoVis Workshop, for which this paper also serves as a technical report.
