Efficient Training of Language Models with Compact and Consistent Next Token Distributions

• URL: http://arxiv.org/abs/2407.02819v1
• Date: Wed, 3 Jul 2024 05:40:41 GMT
• Title: Efficient Training of Language Models with Compact and Consistent Next Token Distributions
• Authors: Ashutosh Sathe, Sunita Sarawagi,
• Abstract summary: We show that we can train better models faster by pre-aggregating the corpus with a collapsed $n$-gram distribution. Our approximation facilitates scalability of gains to larger datasets and models.
• Score: 23.312920633391837
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Maximizing the likelihood of the next token is an established, statistically sound objective for pre-training language models. In this paper we show that we can train better models faster by pre-aggregating the corpus with a collapsed $n$-gram distribution. Previous studies have proposed corpus-level $n$-gram statistics as a regularizer; however, the construction and querying of such $n$-grams, if done naively, prove to be costly and significantly impede training speed, thereby limiting their application in modern large language model pre-training. We introduce an alternative compact representation of the next token distribution that, in expectation, aligns with the complete $n$-gram distribution while markedly reducing variance across mini-batches compared to the standard next-token loss. Empirically, we demonstrate that both the $n$-gram regularized model and our approximation yield substantial improvements in model quality and convergence rate compared to existing methods. Furthermore, our approximation facilitates scalability of gains to larger datasets and models compared to the straightforward $n$-gram regularization method.

Related papers

• The N-Grammys: Accelerating Autoregressive Inference with Learning-Free Batched Speculation [48.52206677611072]
  Speculative decoding aims to speed up autoregressive generation of a language model by verifying in parallel the tokens generated by a smaller draft model. We show that combinations of simple strategies can achieve significant inference speedups over different tasks.
  arXiv  Detail & Related papers  (2024-11-06T09:23:50Z)
• Faster Language Models with Better Multi-Token Prediction Using Tensor Decomposition [5.575078692353885]
  We propose a new model for multi-token prediction in transformers, aiming to enhance sampling efficiency without compromising accuracy. By generalizing it to a rank-$r$ canonical probability decomposition, we develop an improved model that predicts multiple tokens simultaneously.
  arXiv  Detail & Related papers  (2024-10-23T11:06:36Z)
• Exact Byte-Level Probabilities from Tokenized Language Models for FIM-Tasks and Model Ensembles [23.134664392314264]
  Tokenization is associated with many poorly understood shortcomings in language models (LM) This work studies how tokenization impacts model performance by analyzing and comparing models with their byte-level counterparts. We develop a next-byte sampling algorithm that eliminates tokenization bias without requiring further training or optimization.
  arXiv  Detail & Related papers  (2024-10-11T23:30:42Z)
• TokenUnify: Scalable Autoregressive Visual Pre-training with Mixture Token Prediction [61.295716741720284]
  TokenUnify is a novel pretraining method that integrates random token prediction, next-token prediction, and next-all token prediction. Cooperated with TokenUnify, we have assembled a large-scale electron microscopy (EM) image dataset with ultra-high resolution. This dataset includes over 120 million annotated voxels, making it the largest neuron segmentation dataset to date.
  arXiv  Detail & Related papers  (2024-05-27T05:45:51Z)
• Language models scale reliably with over-training and on downstream tasks [121.69867718185125]
  Scaling laws are useful guides for derisking expensive training runs. However, there remain gaps between current studies and how language models are trained. In contrast, scaling laws mostly predict loss on inference, but models are usually compared on downstream task performance.
  arXiv  Detail & Related papers  (2024-03-13T13:54:00Z)
• Compressing Sentence Representation with maximum Coding Rate Reduction [0.0]
  In most natural language inference problems, sentence representation is needed for semantic retrieval tasks. Due to space and time hardware limitations, there is a need to attain comparable results when using the smaller model. We demonstrate that the new language model with reduced complexity and sentence embedding size can achieve comparable results on semantic retrieval benchmarks.
  arXiv  Detail & Related papers  (2023-04-25T09:23:43Z)
• Training Trajectories of Language Models Across Scales [99.38721327771208]
  Scaling up language models has led to unprecedented performance gains. How do language models of different sizes learn during pre-training? Why do larger language models demonstrate more desirable behaviors?
  arXiv  Detail & Related papers  (2022-12-19T19:16:29Z)
• Sparse Upcycling: Training Mixture-of-Experts from Dense Checkpoints [59.39280540478479]
  We propose sparse upcycling -- a simple way to reuse sunk training costs by initializing a sparsely activated Mixture-of-Experts model from a dense checkpoint. We show that sparsely upcycled T5 Base, Large, and XL language models and Vision Transformer Base and Large models, respectively, significantly outperform their dense counterparts on SuperGLUE and ImageNet.
  arXiv  Detail & Related papers  (2022-12-09T18:57:37Z)
• LAVA NAT: A Non-Autoregressive Translation Model with Look-Around Decoding and Vocabulary Attention [54.18121922040521]
  Non-autoregressive translation (NAT) models generate multiple tokens in one forward pass. These NAT models often suffer from the multimodality problem, generating duplicated tokens or missing tokens. We propose two novel methods to address this issue, the Look-Around (LA) strategy and the Vocabulary Attention (VA) mechanism.
  arXiv  Detail & Related papers  (2020-02-08T04:11:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.