SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1801.06146 · 2018

Universal Language Model Fine-tuning for Text Classification

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 5 functions out of this paper's own repositories and ran 2 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
uchange/ulangel reimplementation 2 of 2
prajjwal1/language-modelling reimplementation 0 of 3
FunctionStatusWhere it lives
lower_tw Ran uchange/ulangel/ulangel/data/text_processor.py
code served (permissive licence) · get_code("80418ee432e95278")
toLowercase Ran uchange/ulangel/ulangel/data/text_processor.py
code served (permissive licence) · get_code("6ee9ad2657174087")
get_texts Not yet run prajjwal1/language-modelling/ULMfit.py
pointer only (licence: NONE) · get_code("2a2b16561a736d43")
repackage_var Not yet run prajjwal1/language-modelling/fastai/lm_rnn.py
pointer only (licence: NONE) · get_code("e18c93c24e6c53bf")
seq2seq_reg Not yet run prajjwal1/language-modelling/fastai/lm_rnn.py
pointer only (licence: NONE) · get_code("929d963d360617be")

Repositories linked to this paper

More are linked than shown here.

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Inductive transfer learning has greatly impacted computer vision, but existing approaches in NLP still require task-specific modifications and training from scratch. We propose Universal Language Model Fine-tuning (ULMFiT), an effective transfer learning method that can be applied to any task in NLP, and introduce techniques that are key for fine-tuning a language model. Our method significantly outperforms the state-of-the-art on six text classification tasks, reducing the error by 18-24% on the majority of datasets. Furthermore, with only 100 labeled examples, it matches the performance of training from scratch on 100x more data. We open-source our pretrained models and code.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("1801.06146")
get_code_for_paper("1801.06146")
have("1801.06146")

Connect an agent — have() is free.