26 open source projects by UKPLab

Concatenated Power Mean Embeddings as Universal Cross-Lingual Sentence Representations

Accompanying code for our COLING 2018 paper "Modeling Semantics with Gated Graph Neural Networks for Knowledge Base Question Answering"

Code for paper "End-to-End Non-Factoid Question Answering with an Interactive Visualization of Neural Attention Weights"

The Argument Reasoning Comprehension Task: Source codes & Datasets

Accompanying code for our EMNLP 2017 publication "GraphDocExplore: A Framework for the Experimental Comparison of Graph-based Document Exploration Techniques"

Code and data for ACL2016 article "Which argument is more convincing? Analyzing and predicting convincingness of Web arguments using bidirectional LSTM" by Ivan Habernal and Iryna Gurevych"

E2E NLG Challenge submission

BiLSTM-CNN-CRF architecture for sequence tagging

Hands-on tutorial on deep learning with a special focus on Natural Language Processing (NLP)

Multilingual Sentence & Image Embeddings with BERT

BiLSTM-CNN-CRF architecture for sequence tagging using ELMo representations.

Easy to use, state-of-the-art Neural Machine Translation for 100+ languages

Context-Aware Representations for Knowledge Base Relation Extraction

Code for the paper "Multi-Task Learning for Argumentation Mining in Low-Resource Settings"

No description or website provided.

Accompanying code for our ACL-2017 publication on Neural End-to-End Learning for Computational Argumentation Mining

No description, website, or topics provided.

Investigating Pretrained Language Models for Graph-to-Text Generation

No description, website, or topics provided.

Experimental code for the paper 'Finding Convincing Arguments Using Scalable Bayesian Preference Learning'

No description, website, or topics provided.

A Heterogeneous Benchmark for Information Retrieval. Easy to use, evaluate your models across 15+ diverse IR datasets.

No description, website, or topics provided.

No description, website, or topics provided.

Powerful unsupervised domain adaptation method for dense retrieval. Requires only unlabeled corpus and yields massive improvement: "GPL: Generative Pseudo Labeling for Unsupervised Domain Adaptation of Dense Retrieval" https://arxiv.org/abs/2112.07577