Analyzing The Language of Visual Tokens
Research paper by David M. Chan and collaborators. Published 2024-11-07. Topics: transformer-based models, LLaVA, Chameleon, discrete tokenized representation, image patches.
Resource information
- Authors
- David M. Chan, Rodolfo Corona, Joonyong Park, Cheol Jun Cho, Yutong Bai, Trevor Darrell
- Research topics
- transformer-based models, LLaVA, Chameleon, discrete tokenized representation, image patches, Zipfian distributions, entropy, compression, visual languages, grammatical structures, natural languages, perplexity
- Publication date
- 2024-11-07
- Publication
- arXiv
- Last record update
- Nov 3, 2026