CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering
CRAFT is a query-conditioned pipeline for grounded multi-video question answering over real-world news events. It combines dynamic keyframe selection, per-video ASR with multilingual fallback, and a hybrid critic loop to iteratively verify and repair claims. On the MAGMaR 2026 benchmark, CRAFT achieves the best overall average (0.739), reference recall (0.810), and citation F1 (0.635). It also performs strongly on a WikiVideo conversion (0.823 Avg), demonstrating generalizability. Ablations show atomic claims, ASR, and the critic loop drive main gains.
[2605.19075] CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering
[Submitted on 18 May 2026]
Title:CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering
View a PDF of the paper titled CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering, by Mahesh Bhosale and 5 other authors
View PDF HTML (experimental)
Abstract:Grounded multi-video question answering over real-world news events requires systems to surface query-relevant evidence across heterogeneous video archives while attributing every claim to its supporting source. We introduce CRAFT (Critic-Refined Adaptive Key-Frame Targeting), a query-conditioned pipeline that combines dynamic keyframe selection, per-video ASR with multilingual fallback, and a hybrid critic loop to iteratively verify and repair claims before consolidation. The pipeline integrates UNLI temporal entailment, DeBERTa-v3 cross-claim screening, and a Llama-3.2-3B adjudicator, with a final citation-merging stage that emits each fact once with all supporting source identifiers. On MAGMaR 2026, CRAFT achieves the best overall average (0.739), reference recall (0.810), and citation F1 (0.635). We further evaluate on a MAGMaR-style conversion of WikiVideo with 52 non-overlapping event queries, where CRAFT also performs strongly (0.823 Avg), showing that its claim-centric evidence aggregation generalizes beyond MAGMaR. Ablations show that atomic claims, ASR, and the critic loop drive the main gains over the vanilla query-conditioned baseline. Code and implementation details are publicly available at this https URL.
Comments: Accepted at ACL 2026 Multimodal Augmented Generation via MultimodAl Retrieval Workshop
Subjects:
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
Cite as: arXiv:2605.19075 [cs.CV]
(or arXiv:2605.19075v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2605.19075
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Vishvesh Trivedi [view email] [v1] Mon, 18 May 2026 20:01:05 UTC (137 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled CRAFT: Critic-Refined Adaptive Key-Frame Targeting for Multimodal Video Question Answering, by Mahesh Bhosale and 5 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.CV
new | recent | 2026-05
Change to browse by:
cs cs.AI
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
Loading...
Data provided by:
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)