ViQAgent: Zero-Shot Video Question Answering via Agent with Open-Vocabulary Grounding Validation
This work addresses the challenge of better grounding and decision-making in VideoQA for AI systems, representing an incremental improvement over existing modular and agent-based approaches.
The paper tackles the problem of improving object tracking and alignment in zero-shot Video Question Answering by introducing an LLM-brained agent that combines Chain-of-Thought reasoning with YOLO-World, achieving a new state-of-the-art on benchmarks like NExT-QA, iVQA, and ActivityNet-QA.
Recent advancements in Video Question Answering (VideoQA) have introduced LLM-based agents, modular frameworks, and procedural solutions, yielding promising results. These systems use dynamic agents and memory-based mechanisms to break down complex tasks and refine answers. However, significant improvements remain in tracking objects for grounding over time and decision-making based on reasoning to better align object references with language model outputs, as newer models get better at both tasks. This work presents an LLM-brained agent for zero-shot Video Question Answering (VideoQA) that combines a Chain-of-Thought framework with grounding reasoning alongside YOLO-World to enhance object tracking and alignment. This approach establishes a new state-of-the-art in VideoQA and Video Understanding, showing enhanced performance on NExT-QA, iVQA, and ActivityNet-QA benchmarks. Our framework also enables cross-checking of grounding timeframes, improving accuracy and providing valuable support for verification and increased output reliability across multiple video domains. The code is available at https://github.com/t-montes/viqagent.