GroundedVDR: Grounding Video-to-Web Evidence Dependencies in Video Deep Research
Video large multimodal models have improved at identifying cross-frame visual evidence and conducting open-web retrieval, as reflected in recent video deep research (VDR) benchmarks.
ProofPaper ↗
Key points
- However, existing VDR benchmarks primarily evaluate tool-use trajectories and final responses without verifying whether a model actually uses the temporal video information or the retrieved web evidence.
- To fill this gap, we introduce \method{}, an evidence-grounded benchmark with three features.
- Second, each question carries structured video-to-web evidence annotations that link the relevant video segments to their supporting web evidence.
- Across eight multimodal agent systems, the best reaches 76.0% answer accuracy, yet none recovers the complete required web evidence for more than 8% of questions.
Sources (1)
- [1]GroundedVDR: Grounding Video-to-Web Evidence Dependencies in Video Deep ResearcharXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 18, 07:26 PM
Video large multimodal models have improved at identifying cross-frame visual evidence and conducting open-web retrieval, as reflected in recent video deep research (VDR) benchmarks.
However, existing VDR benchmarks primarily evaluate tool-use trajectories and final responses without verifying whether a model actually uses the temporal video information or the retrieved web evidence.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 18, 2026Comfy-Org/Fibo-2.0
- Oct 17, 2026model : add Kolibri-1 (Aleph-Alpha) support by riteme · Pull Request #30673 · ggml-org/llama.cpp
- Oct 17, 2026Nemotron 3.5 Bolt, Muse Spark Open, and a Fast LFM2.5 24B
- Oct 16, 2026nvidia/NVIDIA-Nemotron-Labs-Dory1.0-4B-MXFP8
- Oct 16, 2026Welcome Beam: a 501B open model you can own, customize, and scale
- Oct 15, 2026Granite Switch 4.1: Chat, Decisions, and Speech from One Endpoint