Video AI Comparison Hub

3 papers - avg viability 6.3

Reference Surfaces

VidVec: Unlocking Video MLLM Embeddings for Video-Text Retrieval(8.0)
VidVec enhances video-text retrieval by leveraging intermediate MLLM layers for state-of-the-art, zero-shot performance.
INFACT: A Diagnostic Benchmark for Induced Faithfulness and Factuality Hallucinations in Video-LLMs(4.0)
INFACT is a diagnostic benchmark designed to evaluate and improve the reliability of Video-LLMs against hallucinations.