Infoblock alpha
cs-cv/vision-language IB-arxiv-2607.28375-fakesv-accuracy measured

Under the FactGuard temporal protocol, HyperClaim achieved 83.7% accuracy on the FakeSV video-misinformation benchmark, outperforming discriminative and reasoning-centric baselines.

Active Evidence state is recorded explicitly and may be revised by a later event.

Observed quantities

accuracy_pct 83.7

01

Primary source

  1. arxiv · 2607.28375 · v1 · abstract
Title
HyperClaim: Fine-Grained Cross-Modal Hypergraph Reasoning for Video Misinformation Detection
Authors
Xiangbo Wang, Jiasheng Zhang, Xingtong Yu, Luoqiang Lei, Delvin Ce Zhang
Version
v1
Check
Support quote found word for word in the arXiv abstract v1, 2026-10-08
Supporting quote from the source
Under the FactGuard temporal protocol, it achieves 83.7%, 82.0%, and 87.3% accuracy on FakeSV, FakeTT, and FakeVV, respectively, outperforming strong discriminative and reasoning-centric baselines

02

Boundary of this block

No additional public limitation was recorded.

03

Connections

Outgoing

No outgoing public edges.

Incoming

No incoming public edges.