TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers/Partially Shuffling the Training Data to Improve Language ...

Partially Shuffling the Training Data to Improve Language Models

Ofir Press

2019-03-11arXiv 2019 3Sentence OrderingLanguage Modelling
PaperPDFCode(official)

Abstract

Although SGD requires shuffling the training data between epochs, currently none of the word-level language modeling systems do this. Naively shuffling all sentences in the training data would not permit the model to learn inter-sentence dependencies. Here we present a method that partially shuffles the training data between epochs. This method makes each batch random, while keeping most sentence ordering intact. It achieves new state of the art results on word-level language modeling on both the Penn Treebank and WikiText-2 datasets.

Results

TaskDatasetMetricValueModel
Language ModellingPenn Treebank (Word Level)Test perplexity52AWD-LSTM-DOC + Partial Shuffle
Language ModellingPenn Treebank (Word Level)Validation perplexity53.79AWD-LSTM-DOC + Partial Shuffle
Language ModellingPenn Treebank (Word Level)Test perplexity53.92AWD-LSTM-MoS + Partial Shuffle
Language ModellingPenn Treebank (Word Level)Validation perplexity55.89AWD-LSTM-MoS + Partial Shuffle
Language ModellingWikiText-2Test perplexity57.85AWD-LSTM-DOC + Partial Shuffle
Language ModellingWikiText-2Validation perplexity60.16AWD-LSTM-DOC + Partial Shuffle
Language ModellingWikiText-2Test perplexity59.98AWD-LSTM-MoS + Partial Shuffle
Language ModellingWikiText-2Validation perplexity62.38AWD-LSTM-MoS + Partial Shuffle

Related Papers

Visual-Language Model Knowledge Distillation Method for Image Quality Assessment2025-07-21Making Language Model a Hierarchical Classifier and Generator2025-07-17VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning2025-07-17The Generative Energy Arena (GEA): Incorporating Energy Awareness in Large Language Model (LLM) Human Evaluations2025-07-17Inverse Reinforcement Learning Meets Large Language Model Post-Training: Basics, Advances, and Opportunities2025-07-17Assay2Mol: large language model-based drug design using BioAssay context2025-07-16Describe Anything Model for Visual Question Answering on Text-rich Images2025-07-16InstructFLIP: Exploring Unified Vision-Language Model for Face Anti-spoofing2025-07-16