Introduces VG-GUIBench benchmark and TASKER keyframe extraction algorithm that improves performance on VideoQA and video-guided agentic tasks.
Adaptive video understanding agent: Enhancing efficiency with dynamic frame sampling and feedback-driven reasoning
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 4years
2026 4roles
background 1polarities
background 1representative citing papers
GridProbe uses posterior probing on a KxK frame grid to adaptively select question-relevant frames, delivering up to 3.36x TFLOPs reduction with accuracy within 1.6 pp of the full-frame baseline on Video-MME-v2.
An edge-cloud pipeline that compresses long video into features, captions, and an entity graph answers streaming-video queries with 87.6% less bandwidth and near cloud-only accuracy.
This is a survey that frames video MLLM research via a human-view formulation of perceptual representations, memory states, reasoning traces, and predictions, then reviews methods, datasets, benchmarks, and open problems.
citing papers explorer
-
Bridging VideoQA and Video-Guided Agentic Tasks via Generalized Keyframe Extraction
Introduces VG-GUIBench benchmark and TASKER keyframe extraction algorithm that improves performance on VideoQA and video-guided agentic tasks.
-
GridProbe: Posterior-Probing for Adaptive Test-Time Compute in Long-Video VLMs
GridProbe uses posterior probing on a KxK frame grid to adaptively select question-relevant frames, delivering up to 3.36x TFLOPs reduction with accuracy within 1.6 pp of the full-frame baseline on Video-MME-v2.
-
CoVStream: Edge-Cloud Collaboration for Understanding of Long Video Streams
An edge-cloud pipeline that compresses long video into features, captions, and an entity graph answers streaming-video queries with 87.6% less bandwidth and near cloud-only accuracy.
-
Watch, Remember, Reason: Human-View Video Understanding with MLLMs
This is a survey that frames video MLLM research via a human-view formulation of perceptual representations, memory states, reasoning traces, and predictions, then reviews methods, datasets, benchmarks, and open problems.