Iryna Gurevych

MDSWriter: Annotation Tool for Creating High-Quality Multi-Document Summarization Corpora

2016

In this paper, we present MDSWriter, a novel open-source annotation tool for creating multi-document summarization corpora. A major innovation of our tool is that we divide the complex summarization task into multiple steps which enables …

Learning to Score System Summaries for Better Content Selection Evaluation.

2017

The evaluation of summaries is a challenging but crucial task of the summarization field. In this work, we propose to learn an automatic scoring metric based on the human judgements available as part of classical …

Automatic Recommendations for Data Coding: A Use Case from Medical and Teacher Education

2018

Research in social sciences and humanities of ten involves analysing data to draw scientific conclusions. This however requires the manual coding of the data, which is highly time-consuming. A use case is the coding of …

Predicting Research Trends From Arxiv

2018 · arXiv (Cornell University)

Knowing trends in research has been a long-standing dream of scientists. Projects on popular research topics often lead to higher acceptance rates at conferences and journals, as well as funding application approvals. Further, knowing future …

Arguments as Social Good: Good Arguments in Times of Crisis

2020 · TUbilio (Technical University of Darmstadt)

We report on a case study about extracting natural language arguments from news media to support decision-making in crises like the Covid-19 pandemic. In particular, we seek to detect the latest pro- and con-arguments and …

Learning to Reason for Text Generation from Scientific Tables

2021 · arXiv (Cornell University)

In this paper, we introduce SciGen, a new challenge dataset for the task of reasoning-aware data-to-text generation consisting of tables from scientific articles and their corresponding descriptions. Describing scientific tables goes beyond the surface realization …

Elastic Weight Removal for Faithful and Abstractive Dialogue Generation

2023 · arXiv (Cornell University)

Ideally, dialogue systems should generate responses that are faithful to the knowledge contained in relevant documents. However, many models generate hallucinated responses instead that contradict it or contain unverifiable information. To mitigate such undesirable behaviour, …

Learning From Free-Text Human Feedback -- Collect New Datasets Or Extend Existing Ones?

2023 · arXiv (Cornell University)

Learning from free-text human feedback is essential for dialog systems, but annotated data is scarce and usually covers only a small fraction of error types known in conversational AI. Instead of collecting and annotating new …

SpaRC and SpaRP: Spatial Reasoning Characterization and Path Generation for Understanding Spatial Reasoning Capability of Large Language Models

2024 · arXiv (Cornell University)

Spatial reasoning is a crucial component of both biological and artificial intelligence. In this work, we present a comprehensive study of the capability of current state-of-the-art large language models (LLMs) on spatial reasoning. To support …

Stepwise Verification and Remediation of Student Reasoning Errors with Large Language Model Tutors

2024 · arXiv (Cornell University)

Large language models (LLMs) present an opportunity to scale high-quality personalized education to all. A promising approach towards this means is to build dialog tutoring models that scaffold students' problem-solving. However, even though existing LLMs …

How Learners Detect Revision Occasions in Texts Labeled as Peer-Written or AI-Generated

2025 · Proceedings.

Although the potential of large language models in education has been widely recognized, research on how learners revise AI-generated content remains limited.The aim of this study was to determine whether learners detect different revision occasions …

An Efficient Quantum Classifier Based on Hamiltonian Representations

2025 · TUbilio (Technical University of Darmstadt)

Quantum machine learning (QML) is a discipline that seeks to transfer the advantages of quantum computing to data-driven tasks. However, many studies rely on toy datasets or heavy feature reduction, raising concerns about their scalability. …

Auditing Language Model Unlearning via Information Decomposition

2026 · TUbilio (Technical University of Darmstadt)

We expose a critical limitation in current approaches to machine unlearning in language models: despite the apparent success of unlearning algorithms, information about the forgotten data remains linearly decodable from internal representations. To systematically assess …

Themis: Training Robust Multilingual Code Reward Models for Flexible Multi-Criteria Scoring

2026 · TUbilio (Technical University of Darmstadt)

Reward models (RMs) have become an indispensable fixture of the language model (LM) post-training playbook, enabling policy alignment and test-time scaling. Research on the application of RMs in code generation, however, has been comparatively sparse, …

Exploiting Debate Portals for Semi-Supervised Argumentation Mining in User-Generated Web Discourse

2015

Analyzing arguments in user-generated Web discourse has recently gained atten-tion in argumentation mining, an evolving field of NLP. Current approaches, which employ fully-supervised machine learn-ing, are usually domain dependent and suffer from the lack of …

Parsing Argumentation Structures in Persuasive Essays

2017 · Computational Linguistics

In this article, we present a novel approach for parsing argumentation structures. We identify argument components using sequence labeling at the token level and apply a new joint model for detecting argumentation structures. The proposed …

Temporal Anchoring of Events for the TimeBank Corpus

2016

Today's extraction of temporal information for events heavily depends on annotated temporal links. These so called TLINKs capture the relation between pairs of event mentions and time expressions. One problem is that the number of …

Optimal Hyperparameters for Deep LSTM-Networks for Sequence Labeling Tasks

2017 · arXiv (Cornell University)

Selecting optimal parameters for a neural network architecture can often make the difference between mediocre and state-of-the-art performance. However, little is published which parameters and design choices should be evaluated or selected making the correct …

Context-Aware Representations for Knowledge Base Relation Extraction

2017

We demonstrate that for sentence-level relation extraction it is beneficial to consider other relations in the sentential context while predicting the target relation. Our architecture uses an LSTM-based encoder to jointly learn representations for all …

A Web-based Tool for the Integrated Annotation of Semantic and Syntactic Structures

2016 · International Conference on Computational Linguistics

We introduce the third major release of WebAnno, a generic web-based annotation tool for distributed teams. New features in this release focus on semantic annotation tasks (e.g. semantic role labelling or event annotation) and allow …

The Argument Reasoning Comprehension Task: Identification and Reconstruction of Implicit Warrants

2018

Ivan Habernal, Henning Wachsmuth, Iryna Gurevych, Benno Stein. Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers). 2018.

ArgumenText: Searching for Arguments in Heterogeneous Sources

2018

Christian Stab, Johannes Daxenberger, Chris Stahlhut, Tristan Miller, Benjamin Schiller, Christopher Tauchmann, Steffen Eger, Iryna Gurevych. Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations. 2018.

The INCEpTION Platform: Machine-Assisted and Knowledge-Oriented Interactive Annotation

2018 · TUbilio (Technical University of Darmstadt)

We introduce INCEpTION, a new annotation platform for tasks including interactive and semantic annotation (e.g., concept linking, fact linking, knowledge base population, semantic frame annotation). These tasks are very time consuming and demanding for annotators, …

UKP-Athene: Multi-Sentence Textual Entailment for Claim Verification

2018

The Fact Extraction and VERification (FEVER) shared task was launched to support the development of systems able to verify claims by extracting supporting or refuting facts from raw text. The shared task organizers provide a …

Ranking Generated Summaries by Correctness: An Interesting but Challenging Application for Natural Language Inference

2019

While recent progress on abstractive summarization has led to remarkably fluent summaries, factual errors in generated summaries still severely limit their use in practice. In this paper, we evaluate summaries produced by state-of-the-art models via …

Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks

2019

Nils Reimers, Iryna Gurevych. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.