Results 241 to 250 of about 152,555 (310)

An Image is Worth 32 Tokens for Reconstruction and Generation

Neural Information Processing Systems
Recent advancements in generative models have highlighted the crucial role of image tokenization in the efficient synthesis of high-resolution images.
Qihang Yu, Mark Weber, Xueqing Deng, Xiaohui Shen, Daniel Cremers, Liang-Chieh Chen   +5 more
semanticscholar   +1 more source

Machine Mental Imagery: Empower Multimodal Reasoning with Latent Visual Tokens

arXiv.org
Vision-language models (VLMs) excel at multimodal understanding, yet their text-only decoding forces them to verbalize visual reasoning, limiting performance on tasks that demand visual imagination.
Zeyuan Yang, Xueyang Yu, Delin Chen, Maohao Shen, Chuang Gan   +4 more
semanticscholar   +1 more source

Tokens valor (security tokens)

2021
Número de páginas ...
openaire   +1 more source

Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research

Annual Meeting of the Association for Computational Linguistics
Information about pretraining corpora used to train the current best-performing language models is seldom discussed: commercial models rarely detail their data, and even open models are often released without accompanying training data or recipes to ...
Luca Soldaini, Rodney Kinney, Akshita Bhagia, Dustin Schwenk, David Atkinson, Russell Authur, Ben Bogin, Khyathi Raghavi Chandu, Jennifer Dumas, Yanai Elazar, Valentin Hofmann, A. Jha, Sachin Kumar, L. Lucy, Xinxi Lyu, Nathan Lambert, Ian Magnusson, Jacob Daniel Morrison, Niklas Muennighoff, Aakanksha Naik, Crystal Nam, Matthew E. Peters, Abhilasha Ravichander, Kyle Richardson, Shannon Zejiang Shen, Emma Strubell, Nishant Subramani, Oyvind Tafjord, Pete Walsh, Luke S. Zettlemoyer, Noah A. Smith, Hanna Hajishirzi, Iz Beltagy, Dirk Groeneveld, Jesse Dodge, Kyle Lo   +35 more
semanticscholar   +1 more source

Token Assorted: Mixing Latent and Text Tokens for Improved Language Model Reasoning

International Conference on Machine Learning
Large Language Models (LLMs) excel at reasoning and planning when trained on chainof-thought (CoT) data, where the step-by-step thought process is explicitly outlined by text tokens.
DiJia Su, Hanlin Zhu, Yingchen Xu, Jiantao Jiao, Yuandong Tian, Qinqing Zheng   +5 more
semanticscholar   +1 more source

Schrödinger's token

Software: Practice and Experience, 2001
AbstractA common problem when writing compilers for programming languages or little, domain‐specific languages is that an input token may have several interpretations, depending on context. Solutions to this problem demand programmer intervention, obfuscate the language's grammar, and may introduce subtle bugs.
John Aycock, R. Nigel Horspool
openaire   +2 more sources

Recent Advances in Discrete Speech Tokens: A Review

IEEE Transactions on Pattern Analysis and Machine Intelligence
The rapid advancement of speech generation technologies in the era of large language models (LLMs) has established discrete speech tokens as a foundational paradigm for speech representation.
Yiwei Guo, Zhihan Li, Hankun Wang, Bohan Li, Chongtian Shao, Hanglei Zhang, Chenpeng Du, Xie Chen, Shujie Liu, Kai Yu   +9 more
semanticscholar   +1 more source

Wait, We Don't Need to "Wait"! Removing Thinking Tokens Improves Reasoning Efficiency

Conference on Empirical Methods in Natural Language Processing
Recent advances in large reasoning models have enabled complex, step-by-step reasoning but often introduce significant overthinking, resulting in verbose and redundant outputs that hinder efficiency.
Chenlong Wang, Yuanning Feng, Dongping Chen, Zhaoyang Chu, Ranjay Krishna, Tianyi Zhou   +5 more
semanticscholar   +1 more source

∞Bench: Extending Long Context Evaluation Beyond 100K Tokens

Volume 1
Processing and reasoning over long contexts is crucial for many practical applications of Large Language Models (LLMs), such as document comprehension and agent construction.
Xinrong Zhang, Yingfa Chen, Shengding Hu, Zihang Xu, Junhao Chen, Moo Khai Hao, Xu Han, Z. Thai, Shuo Wang, Zhiyuan Liu, Maosong Sun   +10 more
semanticscholar   +1 more source

TimeChat-Online: 80% Visual Tokens are Naturally Redundant in Streaming Videos

ACM Multimedia
The rapid growth of online video platforms, particularly live streaming services, has created an urgent need for real-time video understanding systems.
Linli Yao, Yichen Li, Yuancheng Wei, Lei Li, Shuhuai Ren, Yuanxin Liu, Kun Ouyang, Lean Wang, Shicheng Li, Sida Li, Lingpeng Kong, Qi Liu, Yuanxing Zhang, Xu Sun   +13 more
semanticscholar   +1 more source