allenai / allennlp-gallery Goto Github PK
View Code? Open in Web Editor NEWA gallery of projects built with AllenNLP
Home Page: https://gallery.allennlp.org
License: Apache License 2.0
allennlp-gallery's People
Contributors
![dependabot[bot] avatar](https://avatars.githubusercontent.com/in/29110?v=4 "dependabot[bot]")
Stargazers
Watchers
allennlp-gallery's Issues
New model: Neural DRS
Model metadata:
{
"title": "Character-level Representations Improve DRS-based Semantic Parsing Even in the Age of BERT",
"authors": [
{
"name": "Rik van Noord",
"email": "[email protected]",
"affiliation": "University of Groningen",
"google_scholar_author_page": "https://scholar.google.com/citations?user=FB7n4U0AAAAJ",
"s2_author_page": "https://www.semanticscholar.org/author/Rik-van-Noord/66583219"
}, {
"name": "Antonio Toral",
"email": "[email protected]",
"affiliation": "University of Groningen",
"google_scholar_author_page": "https://scholar.google.com/citations?user=YcdNfhcAAAAJ",
"s2_author_page": "https://www.semanticscholar.org/author/Antonio-Toral/144514048"
}, {
"name": "Johan Bos",
"email": "[email protected]",
"affiliation": "University of Groningen",
"google_scholar_author_page": "https://scholar.google.nl/citations?user=Pkj8OSoAAAAJ",
"s2_author_page": "https://www.semanticscholar.org/author/Johan-Bos/3461596"
}
],
"submission_date": "2021-02-02",
"github_link": "https://github.com/RikVN/Neural_DRS",
"paper_link": "https://www.aclweb.org/anthology/2020.emnlp-main.371.pdf",
"allennlp_version": "0.9.0",
"datasets": [
{
"name": "Parallel Meaning Bank",
"link": "https://pmb.let.rug.nl"
}
],
"tags": ["semantic parsing"]
}
Description:
We combine character-level and contextual language model representations to improve performance on Discourse Representation Structure parsing. Character representations can easily be added in a sequence-to-sequence model in either one encoder or as a fully separate encoder, with improvements that are robust to different language models, languages and data sets. For English, these improvements are larger than adding individual sources of linguistic information or adding non-contextual embeddings. A new method of analysis based on semantic tags demonstrates that the character-level representations improve performance across a subset of selected semantic phenomena.
Check and add Multilingual Evidence Retrieval and Fact Verification to Combat Global Disinformation: The Power of Polyglotism
New project: MaChAmp
Project metadata:
{
"title": "Massive Choice, Ample Tasks (MaChAmp):A Toolkit for Multi-task Learning in NLP",
"authors": [
{
"name": "Rob van der Goot",
"email": "[email protected]",
"affiliation": "IT University of Copenhagen"
},
{
"name": "Ahmet Üstün",
"email": "[email protected]",
"affiliation": "University of Groningen"
},
{
"name": "Alan Ramponi",
"email": "[email protected]",
"affiliation": "University of Trento"
},
{
"name": "Ibrahim Sharaf",
"email": "[email protected]",
"affiliation": "Factmata"
},
{
"name": "Barbara Plank",
"email": "[email protected]",
"affiliation": "IT University of Copenhagen"
}
],
"submission_date": "2021-03-30",
"github_link": "https://github.com/machamp-nlp/machamp",
"paper_link": "https://arxiv.org/pdf/2005.14672.pdf",
"allennlp_version": "1.3",
"datasets": [
{
"name": "GLUE",
"link": "https://gluebenchmark.com/"
},
{
"name": "UD",
"link": "https://universaldependencies.org/"
},
{
"name": "WMT14",
"link": "https://nlp.stanford.edu/projects/nmt/"
},
{
"name": "IWSLT15",
"link": "https://nlp.stanford.edu/projects/nmt/"
}
],
"tags": ["MTL", "Multi-task learning", "sequence labeling", "dependency parsing", "text generation", "masked language modeling"]
}
Description:
Transfer learning, particularly approaches that combine multi-task learning with pre-trained contextualized embeddings and fine-tuning, have advanced the field of Natural Language Processing tremendously in recent years. In this paper we present MaChAmp, a toolkit for easy fine-tuning of contextualized embeddings in multi-task settings. The benefits of MaChAmp are its flexible configuration options, and the support of a variety of natural language processing tasks in a uniform toolkit, from text classification and sequence labeling to dependency parsing, masked language modeling, and text generation.
New project: FrisianDutchParsing
Project metadata:
{
"title": "Challenges in Annotating and Parsing Spoken, Code-switched, Frisian-Dutch Data ",
"authors": [
{
"name": "Anouck Braggaar",
"email": "[email protected]",
"affiliation": "University of Groningen",
}, {
"name": "Rob van der Goot",
"email": "[email protected]",
"affiliation": "IT University of Copenhagen"
}
],
"submission_date": "2021-04-01",
"github_link": "https://github.com/Anouck96/ParsingFrisian",
"paper_link": "https://robvanderg.github.io/doc/adapt-nlp2021.2.pdf",
"allennlp_version": "1.3",
"datasets": [
{
"name": "UD_Frisian_Dutch-Fame",
"link": "https://github.com/UniversalDependencies/UD_Frisian_Dutch-Fame/tree/master"
}
],
"tags": ["Universal Dependencies", "UD", "Dependency Parsing", "domain adaptation", "cross-lingual", "cross-domain", "data selection", "Code-switching", "spoken", "Frisian"]
}
Description:
While high performance has been obtained for dependency parsing of high-resource languages, performance for low-resource languages lags behind. In this paper we focus on the parsing of the low-resource language Frisian. We use a sample of code-switched, spontaneously spoken data, which proves to be a challenging setup. We propose to train a parser specifically tailored towards the target domain, by selecting instances from multiple treebanks. Specifically, we use Latent Dirichlet Allocation (LDA), with word and character N-gram features. The best single source treebank (NL ALPINO) resulted in an LAS of 54.7 whereas our data selection outperformed the single best transfer treebank and led to 55.6 LAS on the test data. Additional experiments consisted of removing diacritics from our Frisian data, creating more similar training data by cropping sentences and running our
best model using XLM-R. These experiments did not lead to a better performance.
New project: CodeSwitchNormalizationPOS
Project metadata:
{
"title": "Lexical Normalization for Code-switched Data and its Effect on POS Tagging",
"authors": [
{
"name": "Rob van der Goot",
"email": "[email protected]",
"affiliation": "IT University",
}, {
"name": "Özlem Çetinoğlu",
"email": "[email protected]",
"affiliation": "IMS, University of Stuttgart"
}
],
"submission_date": "2021-04-01",
"github_link": "https://bitbucket.org/robvanderg/csmonoise",
"paper_link": "https://arxiv.org/pdf/2006.01175.pdf",
"allennlp_version": "1.3",
"datasets": [
{
"name": "Turkish-German Code-switching Normalisation Data",
"link": "https://github.com/ozlemcek/TrDeNormData"
},
{
"name": "Indonesian-English Code-switching Normalisation Data",
"link": "https://github.com/seelenbrecher/code-mixed-normalization"
}
],
"tags": ["Lexical normalization", "POS-tagging", "code-switching"]
}
Description:
Lexical normalization, the translation of noncanonical data to standard language, has shown to improve the performance of many natural language processing tasks on social media. Yet, using multiple languages in one utterance, also called code-switching (CS), is frequently overlooked by these normalization systems, despite its common use in social media. In this paper, we propose three normalization models specifically designed to handle codeswitched data which we evaluate for two language pairs: Indonesian-English (Id-En) and Turkish-German (Tr-De). For the latter, we introduce novel normalization layers and their
corresponding language ID and POS tags for the dataset, and evaluate the downstream effect of normalization on POS tagging. Results show that our CS-tailored normalization models outperform Id-En state of the art and Tr-De monolingual models, and lead to 5.4% relative performance increase for POS tagging as compared to unnormalized input.
New project: seq2rel
Hi! Submitting some of our recent work that used AllenNLP for inclusion in the gallery :)
Project metadata:
{
"title": "A sequence-to-sequence approach for document-level relation extraction",
"authors": [
{
"name": "John Giorgi",
"twitter": "@johnmgiorgi",
"email": "[email protected]",
"affiliation": "University of Toronto",
"google_scholar_author_page": "https://scholar.google.ca/citations?user=TNFEhK4AAAAJ&hl=en",
"s2_author_page": "https://www.semanticscholar.org/author/John-Giorgi/37585306"
}, {
"name": "Gary Bader",
"twitter": "@garybader1",
"email": "[email protected]",
"affiliation": "University of Toronto",
"google_scholar_author_page": "https://scholar.google.com/citations?user=22M9eisAAAAJ&hl=en",
},{
"name": "Bo Wang",
"twitter": "@BoWang87",
"email": "[email protected]",
"affiliation": "University of Toronto",
"google_scholar_author_page": "https://scholar.google.com/citations?user=37FDILIAAAAJ&hl=en",
},
],
"submission_date": "2022-04-03",
"github_link": "https://github.com/JohnGiorgi/seq2rel",
"paper_link": "https://aclanthology.org/2022.bionlp-1.2/",
"allennlp_version": "2.9.0",
"datasets": [
{
"name": "CDR",
"link": "https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4860626/"
},
{
"name": "GDA",
"link": "https://link.springer.com/chapter/10.1007/978-3-030-17083-7_17"
},
{
"name": "DGM",
"link": "https://aclanthology.org/N19-1370/"
},
{
"name": "DocRED",
"link": "https://aclanthology.org/P19-1074/"
}
],
"tags": ["named entity recognition", "coreference resolution", "relation extraction", "seq2seq"]
}
Description:
Motivated by the fact that many relations cross the sentence boundary, there has been increasing interest in document-level relation extraction (DocRE). DocRE requires integrating information within and across sentences, capturing complex interactions between mentions of entities. Most existing methods are pipeline-based, requiring entities as input. However, jointly learning to extract entities and relations can improve performance and be more efficient due to shared parameters and training steps. In this paper, we develop a sequence-to-sequence approach, seq2rel, that can learn the subtasks of DocRE (entity extraction, coreference resolution and relation extraction) end-to-end, replacing a pipeline of task-specific components. Using a simple strategy we call entity hinting, we compare our approach to existing pipeline-based methods on several popular biomedical datasets, in some cases exceeding their performance. We also report the first end-to-end results on these datasets for future comparison. Finally, we demonstrate that, under our model, an end-to-end approach outperforms a pipeline-based approach.
New project: SmBop
Project metadata:
{
"title": "SmBoP: Semi-autoregressive Bottom-up Semantic Parsing",
"authors": [
{
"name": "Ohad Rubin",
"email": "[email protected]",
"twitter": "@OhadRubin",
"affiliation": "Tel Aviv University",
"google_scholar_author_page": "https://scholar.google.com/citations?user=4p8NquAAAAAJ",
"s2_author_page": "https://www.semanticscholar.org/author/Ohad-Rubin/2001128224",
}, {
"name": "Jonathan Berant",
"twitter": "@JonathanBerant",
"email": "[email protected]",
"affiliation": "Tel Aviv University, AI2",
"google_scholar_author_page": "https://scholar.google.com/citations?hl=en&user=xCYHonIAAAAJ",
"s2_author_page": :"https://www.semanticscholar.org/author/Jonathan-Berant/1750652",
}
],
"submission_date": "2020-10-23",
"github_link": "https://github.com/OhadRubin/SmBop",
"paper_link": "https://arxiv.org/abs/2010.12412",
"allennlp_version": "2.2.0",
"datasets": [
{
"name": "Spider",
"link": "https://yale-lily.github.io/spider"
}
],
"tags": ["semantic-parsing", "text-to-sql"]
}
Abstract:
The de-facto standard decoding method for semantic parsing in recent years has been to autoregressively decode the abstract syntax tree of the target program using a top-down depth-first traversal. In this work, we propose an alternative approach: a Semi-autoregressive Bottom-up Parser (SmBoP) that constructs at decoding step t the top-K sub-trees of height ≤t. Our parser enjoys several benefits compared to top-down autoregressive parsing. From an efficiency perspective, bottom-up parsing allows to decode all sub-trees of a certain height in parallel, leading to logarithmic runtime complexity rather than linear. From a modeling perspective, a bottom-up parser learns representations for meaningful semantic sub-programs at each step, rather than for semantically-vacuous partial trees. We apply SmBoP on Spider, a challenging zero-shot semantic parsing benchmark, and show that SmBoP leads to a 2.2x speed-up in decoding time and a ∼5x speed-up in training time, compared to a semantic parser that uses autoregressive decoding. SmBoP obtains 71.1 denotation accuracy on Spider, establishing a new state-of-the-art, and 69.5 exact match, comparable to the 69.6 exact match of the autoregressive RAT-SQL+GraPPa.
Other details:
We were accepted to NAACL 2021
Reproducibility details:
We have a Dockerfile that deploys a demo, but it can also be used to train with minor modifications.
We also have a Google Colab demo, but it takes 5 minutes to initialize. Is it possible to host it on https://demo.allennlp.org/ ?
New project: BEESL
Project metadata:
{
"title": "Biomedical Event Extraction as Sequence Labeling ",
"authors": [
{
"name": "Alan Ramponi",
"email": "[email protected]",
"affiliation": "Fondazione the Microsoft Research -- University of Trento. Centre for Computational and Systems Biology (COSBI) and University of Trento",
}, {
"name": "Rob van der Goot",
"email": "[email protected]",
"affiliation": "IT University of Copenhagen"
}
"name": "Rosario Lombardo",
"email": "[email protected]",
"affiliation": "Fondazione the Microsoft Research -- University of Trento. Centre for Computational and Systems Biology (COSBI) and University of Trento"
}, {
"name": "Barbara Plank",
"email": "[email protected]",
"affiliation": "IT University of Copenhagen"
}
],
"submission_date": "2021-04-01",
"github_link": "https://github.com/cosbi-research/beesl",
"paper_link": "https://www.aclweb.org/anthology/2020.emnlp-main.431.pdf",
"allennlp_version": "0.9",
],
"datasets": [
{
"name": "Genia 2011",
"link": "http://2011.bionlp-st.org/home/genia-event-extraction-genia/"
},
"tags": ["Biomedical event extraction", "multi-task learning"]
}
Description:
We introduce Biomedical Event Extraction as Sequence Labeling (BEESL), a joint end to-end neural information extraction model. BEESL recasts the task as sequence labeling, taking advantage of a multi-label aware encoding strategy and jointly modeling the intermediate tasks via multi-task learning. BEESL is fast, accurate, end-to-end, and unlike current methods does not require any external knowledge base or preprocessing tools. BEESL outperforms the current best system (Li et al., 2019) on the Genia 2011 benchmark by 1.57% absolute F1 score reaching 60.22% F1, establishing a new state of the art for the task. Importantly, we also provide first results on biomedical event extraction without gold entity information. Empirical results show that BEESL’s speed and accuracy makes it a viable approach for large-scale real-world scenarios.
New project: MyProject
Project metadata:
{
"title": "Unsupervised Transfer of Semantic Role Models from Verbal to Nominal Domain",
"authors": [
{
"name": "Yanpeng Zhao",
"email": "[email protected]",
"affiliation": "University of Edinburgh",
"google_scholar_author_page": "https://scholar.google.com/citations?user=-T9FigIAAAAJ",
}, {
"name": "Ivan Titov",
"email": "[email protected]",
"affiliation": "University of Edinburgh & University of Amsterdam",
"google_scholar_author_page": "https://scholar.google.com/citations?user=FKUc3vsAAAAJ",
}
],
"submission_date": "2020-09-26",
"github_link": "https://github.com/zhaoyanpeng/srltransfer",
"paper_link": "https://arxiv.org/abs/2005.00278",
"allennlp_version": "0.8.2",
"datasets": [
{
"name": "srl_v2n",
"link": "https://github.com/zhaoyanpeng/srltransfer"
}
],
"tags": ["transfer learning", "semantic role labeling"]
}
Description:
We investigate a transfer scenario where we assume role-annotated data for the source verbal domain but only unlabeled data for the target nominal domain. Our key assumption, enabling the transfer between the two domains, is that selectional preferences of a role (i.e., preferences or constraints on the admissible arguments) do not strongly depend on whether the relation is triggered by a verb or a noun. For example, the same set of arguments can fill the Acquirer role for the verbal predicate acquire and its nominal form acquisition. We approach the transfer task from the variational autoencoding perspective.
New project: VAMPIRE
Project metadata:
{
"title": "Variational Pretraining for Semi-supervised Text Classification",
"authors": [
{
"name": "Suchin Gururangank",
"email": "[email protected]",
"affiliation": "AI2",
"google_scholar_author_page": "https://scholar.google.com/citations?user=CJIKhNIAAAAJ&hl=en",
"s2_author_page": "https://www.semanticscholar.org/author/Suchin-Gururangan/40895369"
}, {
"name": "Tam Dang",
"email": "[email protected]",
"affiliation": "The University of Washington, Seattle"
}, {
"name": "Dallas Card",
"email": " [email protected]",
"affiliation": "Machine Learning Department, Carnegie Mellon University",
"google_scholar_author_page": "https://scholar.google.com/citations?user=qH-rJV8AAAAJ&hl=en"
}, {
"name": "Noah A. Smith",
"email": " [email protected]",
"affiliation": "The University of Washington, Seattle",
"google_scholar_author_page": "https://scholar.google.com/citations?user=TjdFs3EAAAAJ&hl=en"
"s2_author_page": "https://www.semanticscholar.org/author/Noah-A.-Smith/144365875"
}
],
"submission_date": "2012-03-26",
"github_link": "https://github.com/allenai/vampire",
"paper_link": "https://arxiv.org/abs/1906.02242",
"allennlp_version": "0.9.0",
"datasets": [
{
"name": "AG News",
"link": "https://github.com/allenai/vampire/blob/master/scripts/download_ag.sh"
},
{
"name": "IMDB dataset",
"link": "https://github.com/allenai/vampire/blob/master/scripts/download_imdb.sh"
}
],
"tags": ["variational inference", "pretraining", "classification"]
}
Description:
This project is located in AI2 org and is built with AllenNLP. Should be easy to be added.
Add DeCLUTR
Check and add On-Measuring-and-Mitigating-Biased-Inferences-of-Word-Embeddings
New project: DaN+
Project metadata:
{
"title": "DaN+: Danish Nested Named Entities and Lexical Normalization",
"authors": [
{
"name": "Barbara Plank",
"email": "[email protected]",
"affiliation": "IT University of Copenhagen",
}, {
"name": "Kristian Nørgaard Jensen",
"email": "[email protected]",
"affiliation": "IT University of Copenhagen"
}, {
"name": "Rob van der Goot",
"email": "[email protected]",
"affiliation": "IT University of Copenhagen"
} ],
"submission_date": "01-04-2021",
"github_link": "https://github.com/bplank/DaNplus",
"paper_link": "https://www.aclweb.org/anthology/2020.coling-main.583.pdf",
"allennlp_version": "1.1",
"datasets": [
{
"name": "DaN+",
"link": "https://github.com/bplank/DaNplus"
}
],
"tags": ["named entity recognition", "named entity detection", "lexical normalization", "domain adaptation", "Danish"]
}
Description:
This paper introduces DAN+, a new multi-domain corpus and annotation guidelines for Danish nested named entities (NEs) and lexical normalization to support research on cross-lingual cross-domain learning for a less-resourced language. We empirically assess three strategies to model the two-layer Named Entity Recognition (NER) task. We compare transfer capabilities from German versus in-language annotation from scratch. We examine language-specific versus multilingual BERT, and study the effect of lexical normalization on NER. Our results show that 1) the most robust strategy is multi-task learning which is rivaled by multi-label decoding, 2) BERT-based NER models are sensitive to domain shifts, and 3) in-language BERT and lexical normalization are the most beneficial on the least canonical data. Our results also show that an out-of-domain setup remains challenging, while performance on news plateaus quickly. This highlights the importance of cross-domain evaluation of cross-lingual transfer.
Add truecaser?
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.