Summary of Temporalbench: Benchmarking Fine-grained Temporal Understanding For Multimodal Video Models, by Mu Cai et al.
TemporalBench: Benchmarking Fine-grained Temporal Understanding for Multimodal Video Models
by Mu Cai, Reuben Tan, Jianrui Zhang, Bocheng Zou, Kai Zhang, Feng Yao, Fangrui Zhu, Jing Gu, Yiwu Zhong, Yuzhang Shang, Yao Dou, Jaden Park, Jianfeng Gao, Yong Jae Lee, Jianwei Yang
First submitted to arxiv on: 14 Oct 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper introduces TemporalBench, a new benchmark for evaluating fine-grained temporal understanding in videos. The existing video benchmarks are inadequate for assessing temporal comprehension due to the lack of fine-grained annotations. TemporalBench consists of over 10K question-answer pairs derived from high-quality human annotations detailing the temporal dynamics in video clips. This benchmark enables evaluations on various tasks, including video question answering and captioning, as well as different models such as multimodal video embedding models and text generation models. State-of-the-art models like GPT-4o achieve only 38.5% question answering accuracy on TemporalBench, demonstrating a significant gap between humans and AI in temporal understanding. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper creates a new way to test how well computers understand videos by looking at the small details that happen over time. Right now, most video tests are like still image tests because they don’t show the tiny changes that happen in a video. This paper makes a special test called TemporalBench that shows these small details and lets us see if computers can understand them too. Computers aren’t very good at this yet – even the best ones only get 38.5% of questions right! This means there’s still a lot to learn for computers. |
Keywords
» Artificial intelligence » Embedding » Gpt » Question answering » Text generation
