paper-with-me

Papers

Language Modeling Teaches You More Syntax than Translation Does: Lessons Learned Through Auxiliary Task Analysis

2018-09-26 · Kelly W. Zhang, Samuel R. Bowman

Recent work using auxiliary prediction task classifiers to investigate the properties of LSTM representations has begun to shed light on why pretrained representations, like ELMo (Peters et al., 2018) and CoVe (McCann et al., 2017), are so beneficial for neural language understanding models. We still, though, do not yet have a clear understanding of how the choice of pretraining objective affects the type of linguistic information that models learn. With this in mind, we compare four objectives---language modeling, translation, skip-thought, and autoencoding---on their ability to induce syntactic and part-of-speech information. We make a fair comparison between the tasks by holding constant the quantity and genre of the training data, as well as the LSTM architecture. We find that representations from language models consistently perform best on our syntactic auxiliary prediction tasks, even when trained on relatively small amounts of data. These results suggest that language modeling may be the best data-rich pretraining task for transfer learning applications requiring syntactic information. We also find that the representations from randomly-initialized, frozen LSTMs perform strikingly well on our syntactic auxiliary tasks, but this effect disappears when the amount of training data for the auxiliary tasks is reduced.

📄 PDF Abstract BibTeX arXiv:1809.10040

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage ModellingTransfer LearningTranslation

Methods 이 논문이 사용한 방법론

Sigmoid Activation 설명 없음
Tanh Activation 설명 없음
GloVe GloVe Embeddings are a type of word embedding that encode the co-occurrence probability ratio between two words as vector differences. GloVe uses a weighted least squares…
Location-based Attention 설명 없음
Seq2Seq Seq2Seq, or Sequence To Sequence, is a model used in sequence prediction tasks, such as language modelling and machine translation. The idea is to use one…
CoVe CoVe, or Contextualized Word Vectors, uses a deep LSTM encoder from an attentional sequence-to-sequence model trained for machine…
BiLSTM A Bidirectional LSTM, or biLSTM, is a sequence processing model that consists of two LSTMs: one taking the input in a forward direction, and the other in a backwards…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

Predicting Target Language CCG Supertags Improves Neural Machine Translation

2017-02-03 · WS 2017 9 · Maria Nadejde, Siva Reddy, Rico Sennrich, Tomasz Dwojak 외

Neural machine translation (NMT) models are able to partially learn syntactic information from sequential lexical information. Still, some complex syntactic phenomena such as prepositional phrase attachment are poorly mo…

DecoderMachine TranslationNMTPrepositional Phrase Attachment+1

Explicitly Modeling Syntax in Language Models with Incremental Parsing and a Dynamic Oracle

2020-10-21 · NAACL 2021 4 · Yikang Shen, Shawn Tan, Alessandro Sordoni, Siva Reddy 외

Syntax is fundamental to our thinking about language. Failing to capture the structure of input language could lead to generalization problems and over-parametrization. In the present work, we propose a new syntax-aware …

Language ModelingLanguage Modelling

Syntax Without Semantics: Teaching Large Language Models to Code in an Unseen Language

2026-05-15 · Vinayshekhar Bannihatti Kumar, Disha Makhija, Manoj Ghuhan Arivazhagan, Rashmi Gangadharaiah arxiv

Large language models (LLMs) achieve high pass rates on code generation benchmarks, yet whether they can transfer this ability to languages absent from pretraining remains poorly understood. We introduce PyLang, a minima…

Multi-Task LearningCode Generation

Language Modeling Teaches You More than Translation Does: Lessons Learned Through Auxiliary Syntactic Task Analysis

2018-11-01 · WS 2018 11 · Kelly Zhang, Samuel Bowman

Recently, researchers have found that deep LSTMs trained on tasks like machine translation learn substantial syntactic and semantic information about their input sentences, including part-of-speech. These findings begin …

CCG SupertaggingLanguage ModelingLanguage ModellingMachine Translation+3

CodeAlchemy: Synthetic Code Rewriting at Scale

2026-06-08 · Ankit Gupta, Aditya Prasad, Rameswar Panda arxiv

Pre-training on raw code teaches syntax but provides sparse signal for diverse real-world task formats. While synthetic data has proven transformative for language models, code remains largely unexplored beyond limited q…

Synthetic Data Generation