Papers by Nils Feldhus
Cross-Refine: Improving Natural Language Explanation Generation by Learning in Tandem (2025.coling-main)
Copied to clipboard
| Challenge: | Natural language explanations (NLEs) are vital for elucidating the reasoning behind large language model (LLM) decisions. |
| Approach: | They propose a role-modeling approach that employs two LLMs as generator and critic to generate and refine NLEs. |
| Outcome: | The proposed model outperforms self-refine and can perform with less powerful LLMs. |
CoXQL: A Dataset for Parsing Explanation Requests in Conversational XAI Systems (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing systems based on large language models (LLMs) are more precise and reliable in identifying users’ intentions, but the recognition of intents still presents a challenge in the case of ConvXAI, since little training data exist and the domain is highly specific. |
| Approach: | They propose to use a dataset in the NLP domain for user intent recognition in ConvXAI to improve parsing performance. |
| Outcome: | The proposed system outperforms existing methods and improves on existing ones. |
Parallel Universes, Parallel Languages: A Comprehensive Study on LLM-based Multilingual Counterfactual Example Generation (2026.acl-long)
Copied to clipboard
Qianli Wang, Van Bach Nguyen, Yihong Liu, Fedor Splitt, Nils Feldhus, Christin Seifert, Hinrich Schuetze, Sebastian Möller, Vera Schmitt
| Challenge: | Large language models excel at generating English counterfactuals but their effectiveness in generating multilingual counterfacts remains unclear. |
| Approach: | They conduct automatic evaluations on both directly generated and derived counterfactuals in six languages and find that cross-lingual perturbations follow common strategic principles. |
| Outcome: | The proposed models show that translation-based counterfactuals offer higher validity than their directly generated counterparts, but still fall short of matching the quality of the original English counterf actuals. |
Simplifying Outcomes of Language Model Component Analyses with ELIA (2026.eacl-demo)
Copied to clipboard
| Challenge: | ELIA is an interactive web application that simplifies the outputs of various language model component analyses for a broader audience. |
| Approach: | They propose to use a vision-language model to automatically generate natural language explanations for the complex visualizations produced by these methods. |
| Outcome: | The proposed system integrates three key techniques and generates natural language explanations for complex visualizations. |
European Language Grid: A Joint Platform for the European Language Technology Community (2021.eacl-demos)
Copied to clipboard
Georg Rehm, Stelios Piperidis, Kalina Bontcheva, Jan Hajic, Victoria Arranz, Andrejs Vasiļjevs, Gerhard Backfried, Jose Manuel Gomez-Perez, Ulrich Germann, Rémi Calizzano, Nils Feldhus, Stefanie Hegele, Florian Kintzel, Katrin Marheinecke, Julian Moreno-Schneider, Dimitris Galanis, Penny Labropoulou, Miltos Deligiannis, Katerina Gkirtzou, Athanasia Kolovou, Dimitris Gkoumas, Leon Voukoutis, Ian Roberts, Jana Hamrlova, Dusan Varis, Lukas Kacena, Khalid Choukri, Valérie Mapelli, Mickaël Rigault, Julija Melnika, Miro Janosik, Katja Prinz, Andres Garcia-Silva, Cristian Berrio, Ondrej Klejch, Steve Renals
| Challenge: | Europe is a multilingual society, in which dozens of languages are spoken. |
| Approach: | They describe the European Language Grid, which is targeted to evolve into the primary platform and marketplace for LT in Europe by providing one umbrella platform for the European LT landscape. |
| Outcome: | The European Language Grid (ELG) will provide access to 1300 services for all European languages as well as thousands of data sets. |
FitCF: A Framework for Automatic Feature Importance-guided Counterfactual Example Generation (2025.findings-acl)
Copied to clipboard
Qianli Wang, Nils Feldhus, Simon Ostermann, Luis Felipe Villa-Arenas, Sebastian Möller, Vera Schmitt
| Challenge: | Existing frameworks for counterfactual examples are lacking for many tasks. |
| Approach: | They propose a faithful approach for leveraging important words from feature attribution methods to generate counterfactual examples in a zero-shot setting. |
| Outcome: | The proposed framework outperforms state-of-the-art frameworks on many tasks. |
An Annotated Corpus of Textual Explanations for Clinical Decision Support (2022.lrec-1)
Copied to clipboard
Roland Roller, Aljoscha Burchardt, Nils Feldhus, Laura Seiffe, Klemens Budde, Simon Ronicke, Bilgin Osmanodja
| Challenge: | In recent years, machine learning for clinical decision support has gained more and more attention. |
| Approach: | They propose to use XAI to provide an explanation of a model's decision making process by constructing a corpus of sentences that are annotated with different semantic layers. |
| Outcome: | The proposed models outperform physicians on very specific, narrow tasks or can help physicians to work more efficiently. |
Inseq: An Interpretability Toolkit for Sequence Generation Models (2023.acl-demo)
Copied to clipboard
| Challenge: | Recent studies focused on classification tasks while largely overlooking generation settings due to a lack of dedicated tools. |
| Approach: | They propose to use Inseq to democratize access to interpretability analyses of sequence generation models by enabling intuitive extraction of models’ internal information and feature importance scores for popular decoder-only and encoder-decoder Transformers architectures. |
| Outcome: | The proposed library can extract models’ internal information and feature importance scores for popular decoder-only and encoder-decoder Transformers architectures. |
Thermostat: A Large Collection of NLP Model Explanations and Analysis Tools (2021.emnlp-demo)
Copied to clipboard
| Challenge: | Arras et al. (2016): explainability methods are perceived as opaque due to their complexity. |
| Approach: | They propose to use model explanations and analysis tools to facilitate research . they use a dataset that took 10k GPU hours to compile and analyse . |
| Outcome: | Thermostat allows easy access to over 200k explanations for state-of-the-art models . dataset took over 10k GPU hours (> one year) to compile; saves time . |
Persona Prompting as a Lens on LLM Social Reasoning (2026.eacl-long)
Copied to clipboard
Jing Yang, Moritz Hechtbauer, Elisabeth Khalilov, Evelyn Luise Brinkmann, Vera Schmitt, Nils Feldhus
| Challenge: | Persona prompting (PP) is increasingly used to steer large language models towards user-specific generation, but its effect on rationales remains underexplored. |
| Approach: | They examine how LLM-generated rationales vary when conditioned on different demographic personas . they use word-level rationale annotations to measure agreement with human annotations based on PP . |
| Outcome: | The proposed model improves classification on the most subjective task, but fails to align with real-world demographic counterparts. |
Infherno: End-to-end Agent-based FHIR Resource Synthesis from Free-form Clinical Notes (2026.eacl-demo)
Copied to clipboard
| Challenge: | HL7 FHIR format is a desirable format for clinical data integration and healthcare services. |
| Approach: | They propose an end-to-end framework that adheres to the HL7 FHIR document schema . it uses LLM agents, code execution, and healthcare terminology database tools . |
| Outcome: | The proposed framework adheres to the HL7 FHIR document schema and competes well with a human baseline in predicting FHIr resources from unstructured text. |
Multilingual Datasets for Custom Input Extraction and Explanation Requests Parsing in Conversational XAI Systems (2025.findings-emnlp)
Copied to clipboard
Qianli Wang, Tatiana Anikina, Nils Feldhus, Simon Ostermann, Fedor Splitt, Jiaao Li, Yoana Tsoneva, Sebastian Möller, Vera Schmitt
| Challenge: | Current ConvXAI systems are based on intent recognition to accurately identify the user’s desired intention and map it to an explainability method. |
| Approach: | They propose a multilingual extension of the CoXQL dataset spanning five typologically diverse languages, including one low-resource language. |
| Outcome: | The proposed model enables multilingual generalization in a multilingual dataset spanning five typologically diverse languages, including one low-resource language. |
InterroLang: Exploring NLP Models and Datasets through Dialogue-based Explanations (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Recent work on NLP explainability methods lacks a dialogue-based interpretability framework that can convey faithful explanations in human-understandable terms. |
| Approach: | They adapt the conversational explanation framework TalkToModel to the NLP domain and add new NLP-specific operations such as free-text rationalization to illustrate its generalizability. |
| Outcome: | The proposed framework can be used to explain models on three NLP tasks and is generalizable to different datasets, use cases and models. |