awesome-nlp

canonical archive of papers from ACL, EMNLP, NAACL, EACL, COLING, and related venues.

the rolling review process feeding ACL-affiliated venues.

An NLP research library, built on PyTorch, for developing state-of-the-art deep learning models on a wide variety of linguistic tasks.

NLP and ML suite covers most common tasks like NER, tagging, and sentiment analysis

Free End-to-End No-Code platform for text annotation and DL model training/tuning. Out-of-the-box support for Named Entity Recognition, Classification, Relation extraction and Assertion Status Spark NLP models. Unlimited support for users, teams, projects, documents. Not FOSS.

open-source platform for collecting human feedback, building NLP and LLM datasets, and curating preference data.

LLMs for context-aware translation.

decoder-only LLMs with a novel error-rate adaptation schedule set new SOTA on BEA-test grammatical error correction.

retrieval-augmented LM for few-shot QA.

transformer; reset the field.

LLMs vs fine-tuned summarizers.

clustering-based topic modeling on top of contextual embeddings; common modern default.

CS 685, UMass Amherst CS

Free online course covering text processing, large-scale data analysis, processing pipelines, and training neural network models for custom NLP tasks.

a16z AI playbook is a great link to forward to your managers or content for your presentations

free course on LLMs, embeddings, semantic search, and NLP applications.

suite of open LLMs (4B-14B) continued-pretrained on 26B tokens across 20 African languages with a comprehensive empirical study of data mixing.

models strategically complying during training.

on-device 3B model using KV-cache sharing and 2-bit QAT for 37.5% cache memory reduction without accuracy loss.

what BERT learns about language.

claim-level calibration analysis for long-form generation; models are substantially worse-calibrated on extended outputs than on single claims.

activation-aware weight quantization.

tools, datasets, and models across 22 Indic languages.

instruction-tuned Hindi LLM.

encoder-only Portuguese LMs for both PT-PT and PT-BR.

Arabic-first foundation models.

dependency parser for Dutch (also does POS tagging and lemmatization).

Arabic BERT family.

generative AI across modalities.

general-purpose large language model resources.

production ML, including LLM serving.

prompting techniques and template libraries.

retrieval-augmented generation systems and tooling.

2T-token open-license multilingual corpus.

6.3T tokens across 167 languages.

3T-token open pretraining corpus with documented filtering pipeline.

15T-token cleaned web corpus; FineWeb-Edu filters for educational quality.

data repository for pretrained NLP models and NLP corpora.