AI Breakdown
arxiv preprint - CinePile: A Long Video Question Answering Dataset and Benchmark
30 May 2024
In this episode, we discuss CinePile: A Long Video Question Answering Dataset and Benchmark by Ruchit Rawal, Khalid Saifullah, Ronen Basri, David Jacobs, Gowthami Somepalli, Tom Goldstein. CinePile is a new dataset and benchmark designed for authentic long-form video understanding, addressing the limitations of current datasets. It comprises 305,000 multiple-choice questions (MCQs) spanning various visual and multimodal aspects. The evaluation of recent state-of-the-art video-centric language models (LLMs) shows a significant gap between machine and human performance in these complex tasks.
No persons identified in this episode.
This episode hasn't been transcribed yet
Help us prioritize this episode for transcription by upvoting it.
Popular episodes get transcribed faster
Other recent transcribed episodes
Transcribed and ready to explore now
#2425 - Ethan Hawke
11 Dec 2025
The Joe Rogan Experience
SpaceX Said to Pursue 2026 IPO
10 Dec 2025
Bloomberg Tech
Don’t Call It a Comeback
10 Dec 2025
Motley Fool Money
Japan Claims AGI, Pentagon Adopts Gemini, and MIT Designs New Medicines
10 Dec 2025
The Daily AI Show
Eric Larsen on the emergence and potential of AI in healthcare
10 Dec 2025
McKinsey on Healthcare
What it will take for AI to scale (energy, compute, talent)
10 Dec 2025
Azeem Azhar's Exponential View