Papers by Dietrich Klakow

Preventing Author Profiling through Zero-Shot Multilingual Back-Translation (2021.emnlp-main)

Challenge:	Large Language Models (LLMs) are increasingly employed in high-stakes decision-making tasks such as loan approvals.
Approach:	They evaluate the performance and fairness of LLMs on serialized loan approval datasets from Ghana, Germany, and the United States.
Outcome:	The model’s zero-shot and in-context learning (ICL) capabilities are evaluated on loan approval datasets from Ghana, Germany, and the United States.

Copied to clipboard

David Adelani, Miaoran Zhang, Xiaoyu Shen, Ali Davody, Thomas Kleinbauer, Dietrich Klakow

PricingLogic: Evaluating LLMs Reasoning on Complex Tourism Pricing Tasks (2025.emnlp-main)

Challenge:	Documents as short as a single sentence may reveal sensitive information about authors . style transfer is effective but a number of current methods cause a drop in down-stream utility .
Approach:	They propose a method to remove sensitive information from documents by multilingual back-translation using off-the-shelf translation models.
Outcome:	The proposed method lowers adversarial gender and race prediction by 22% while retaining 95% of original utility on downstream tasks.

Copied to clipboard

Yunuo Liu, Dawei Zhu, Zena Al-Khalili, Dai Cheng, Yanjun Chen, Dietrich Klakow, Wei Zhang, Xiaoyu Shen

Reasoning

Challenge:	Recent advances in Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse domains, such as code generation, mathematical problem-solving, and general-purpose human instruction following.
Approach:	They propose to use large language models to process questions expressed in natural language to automate tourism-booking prices when multiple, overlapping farerules apply.
Outcome:	The proposed model can automate tourism-booking prices when multiple, overlapping farerules apply.

Unveiling the Key Factors for Distilling Chain-of-Thought Reasoning (2025.findings-acl)

Copied to clipboard

Xinghao Chen, Zhijing Sun, Guo Wenjin, Miaoran Zhang, Yanjun Chen, Yirong Sun, Hui Su, Yijie Pan, Dietrich Klakow, Wenjie Li, Xiaoyu Shen

From Insights to Actions: The Impact of Interpretability and Analysis Research on NLP (2024.emnlp-main)

Challenge:	Large Language Models (LLMs) excel in reasoning tasks through Chain-of-Thought prompting.
Approach:	They examine the factors influencing CoT distillation including granularity, format and teacher model.
Outcome:	The proposed model is based on four teacher models and seven student models across seven mathematical and commonsense reasoning datasets.

Copied to clipboard

Marius Mosbach, Vagrant Gautam, Tomás Vergara Browne, Dietrich Klakow, Mor Geva

NEXUS Network: Connecting the Preceding and the Following in Dialogue Generation (D18-1)

Challenge:	Interpretability and analysis (IA) research is a growing subfield within NLP . a criticism of this work is that it lacks actionable insights and therefore has little impact on NLP.
Approach:	They propose to quantify the impact of interpretation and analysis research on NLP . they use citation graphs and a survey to find out what is missing in IA research .
Outcome:	The proposed study shows that IA research is well-cited outside of IA and central in the NLP citation graph.

Copied to clipboard

Xiaoyu Shen, Hui Su, Wenjie Li, Dietrich Klakow

Dialogue Systems & Conversational Agents

Challenge:	Sequence-to-Sequence models favor short generic responses . however, the model is not suitable for modeling dialogues .
Approach:	They propose a model that connects preceding and following conversations to a prior distribution to avoid non-differentiability of discrete natural language tokens.
Outcome:	The proposed model is highly efficient in learning the backbone of human-computer communications, but favors short generic responses.

Exploring the Effectiveness and Consistency of Task Selection in Intermediate-Task Transfer Learning (2024.acl-srw)

Copied to clipboard

Pin-Jie Lin, Miaoran Zhang, Marius Mosbach, Dietrich Klakow

HUMAN: Hierarchical Universal Modular ANnotator (2020.emnlp-demos)

Challenge:	Identifying beneficial tasks to transfer from is a critical step toward successful intermediate-task transfer learning.
Approach:	They propose a method that measures pairwise token similarity using maximum inner product search to improve task prediction.
Outcome:	The proposed method improves task prediction scores from 2.59% to 3.96% for tasks requiring reasoning abilities, but not for reasoning abilities.

Copied to clipboard

Moritz Wolf, Dana Ruiter, Ashwin Geet D’Sa, Liane Reiners, Jan Alexandersson, Dietrich Klakow

Challenge:	HUMAN is a web-based annotation tool that covers a variety of annotation tasks on textual and image data.
Approach:	They propose a web-based annotation tool that covers a variety of annotation tasks on textual and image data.
Outcome:	HUMAN covers a variety of annotation tasks on textual and image data and uses an internal deterministic state machine to chain different tasks in an interdependent manner.

Weaker Than You Think: A Critical Look at Weakly Supervised Learning (2023.acl-long)

Copied to clipboard

Dawei Zhu, Xiaoyu Shen, Marius Mosbach, Andreas Stephan, Dietrich Klakow

A Preference-driven Paradigm for Enhanced Translation with Large Language Models (2024.naacl-long)

Challenge:	Weakly supervised learning is a popular approach for training machine learning models in low-resource settings.
Approach:	They propose to use weakly supervised learning to train models with noisy labels from weak sources instead of collecting expensive human annotations.
Outcome:	The proposed methods outperform weakly supervised methods on various NLP datasets and tasks on the test sets.

Copied to clipboard

Dawei Zhu, Sony Trenous, Xiaoyu Shen, Dietrich Klakow, Bill Byrne, Eva Hasler

Meta Self-Refinement for Robust Learning with Weak Supervision (2023.eacl-main)

Challenge:	Recent research shows that large language models (LLMs) can achieve remarkable translation performance through supervised fine-tuning (SFT) however, SFT simply instructs the model to imitate reference translations token by token, making it vulnerable to the noise present in the data.
Approach:	They propose a preference-based approach to supervised fine-tuning that trains the model to imitate reference translations token by token, making it vulnerable to noise.
Outcome:	The proposed approach overcomes the plateau associated with imitation-based SFT and is more resilient in the absence of gold translations.

Copied to clipboard

Dawei Zhu, Xiaoyu Shen, Michael Hedderich, Dietrich Klakow

A Lightweight Method to Generate Unanswerable Questions in English (2023.findings-emnlp)

Challenge:	Recent methods leverage self-training to build noise-resistant models . however, the teacher trained under weak supervision may have fitted a substantial amount of noise and therefore produce incorrect pseudo-labels.
Approach:	They propose a framework that encourages teacher to refine its pseudo-labels to effectively combat label noise from weak supervision.
Outcome:	The proposed framework outperforms state-of-the-art methods by 11.4% in accuracy and 9.26% in F1 score on eight NLP benchmarks.

Copied to clipboard

Vagrant Gautam, Miaoran Zhang, Dietrich Klakow

Neural Data-to-Text Generation via Jointly Learning the Segmentation and Correspondence (2020.acl-main)

Challenge:	Existing approaches to build robust question answering models are too complex . antonym and entity swaps on answerable questions are used to build models .
Approach:	They propose a method for performing antonym and entity swaps on unanswerable questions.
Outcome:	The proposed method outperforms the previous state-of-the-art and has higher human-judged relatedness and readability.

Copied to clipboard

Xiaoyu Shen, Ernie Chang, Hui Su, Cheng Niu, Dietrich Klakow

AFRIDOC-MT: Document-level MT Corpus for African Languages (2025.emnlp-main)

Challenge:	Recent neural attention models conflate all steps into a single end-to-end system and simplify training process.
Approach:	They propose to explicitly segment target text into fragment units and align them with their data correspondences.
Outcome:	The proposed model outperforms neural attention models on E2E and WebNLG benchmarks.

Copied to clipboard

Jesujoba Oluwadara Alabi, Israel Abebe Azime, Miaoran Zhang, Cristina España-Bonet, Rachel Bawden, Dawei Zhu, David Ifeoluwa Adelani, Clement Oyeleke Odoje, Idris Akinade, Iffat Maab, Davis David, Shamsuddeen Hassan Muhammad, Neo Putini, David O. Ademuyiwa, Andrew Caines, Dietrich Klakow

Adapting Pre-trained Language Models to African Languages via Multilingual Adaptive Fine-Tuning (2022.coling-1)

Challenge:	AFRIDOC-MT is a document-level multi-parallel translation dataset covering five languages . AFRITIC-MT models perform better on sentences than general-purpose LLMs .
Approach:	They propose a document-level multi-parallel translation dataset covering English and five African languages.
Outcome:	The proposed dataset covers 334 health and 271 information technology news documents . it shows that NLLB-200 achieves the best average performance among standard models .

Copied to clipboard

Jesujoba O. Alabi, David Ifeoluwa Adelani, Marius Mosbach, Dietrich Klakow

Select and Attend: Towards Controllable Content Selection in Text Generation (D19-1)

Challenge:	Multilingual pre-trained language models have shown impressive performance on several downstream tasks for both high-resourced and low-resource languages.
Approach:	They propose to apply multilingual adaptive fine-tuning to 17 most-resourced African languages and three other high-resource languages to encourage cross-lingual transfer learning.
Outcome:	The proposed approach is competitive to LAFT on individual languages while requiring significantly less disk space.

Copied to clipboard

Xiaoyu Shen, Jun Suzuki, Kentaro Inui, Hui Su, Dietrich Klakow, Satoshi Sekine

To Share or not to Share: Predicting Sets of Sources for Model Transfer Learning (2021.emnlp-main)

Challenge:	Recent neural network models conflate content selection and surface realization into a black-box architecture, resulting in content to be described in text cannot be explicitly controlled.
Approach:	They propose to decouple content selection from the decoder to allow finer-grained control over the generation.
Outcome:	The proposed model can be trained end-to-end without human annotations and achieves promising results in data-totext and headline generation tasks.

Copied to clipboard

Lukas Lange, Jannik Strötgen, Heike Adel, Dietrich Klakow

Adapting Language Models When Training on Privacy-Transformed Data (2022.lrec-1)

Challenge:	Existing methods to select transfer sources are limited by text and task similarity, which limits their application in transfer settings where both the task and the text domain change.
Approach:	They propose a model similarity measure that represents text and task similarity jointly to automatically determine which and how many sources to exploit.
Outcome:	The proposed approach improves performance by 24 F1 points for predicting promising sources across domains and tasks with similar models.

Copied to clipboard

Tugtekin Turan, Dietrich Klakow, Emmanuel Vincent, Denis Jouvet

Improving Latent Alignment in Text Summarization by Generalizing the Pointer Generator (D19-1)

Challenge:	Using data sanitization methods to remove personal information from spoken messages is not effective because privacy-transformed data is unlikely to match the test distribution.
Approach:	They propose to use a data sanitization approach to remove personal information from spoken messages by replacing named entities with other words from the same class.
Outcome:	The proposed approach removes personal information from the spoken messages using an automatic named entity recognition method.

Copied to clipboard

Xiaoyu Shen, Yang Zhao, Hui Su, Dietrich Klakow

Aligned Probing: Relating Toxic Behavior and Model Internals (2026.tacl-1)

Challenge:	Modern pointer generators only capture exact word matches, ignoring possible inflections or abstractions, which restricts its power of capturing richer latent alignment.
Approach:	They propose a pointer generator architecture that allows the model to "edit" pointed tokens instead of always copying them.
Outcome:	The proposed model captures more latent alignment relations than exact word matches and generates higher-quality summaries validated by both qualitative and quantitative evaluations.

Copied to clipboard

Andreas Waldis, Vagrant Gautam, Anne Lauscher, Dietrich Klakow, Iryna Gurevych

EthioLLM: Multilingual Large Language Models for Ethiopian Languages with Task Evaluation (2024.lrec-main)

Challenge:	Language models (LMs) may produce toxic text that contains hate speech, insults, or vulgarity, even when prompted with innocuous text.
Approach:	They propose an interpretability framework that aligns the behavior of language models based on their outputs and internal representations.
Outcome:	The proposed framework bridges behavioral and internal perspectives for toxicity for the first time.

Copied to clipboard

Atnafu Lambebo Tonja, Israel Abebe Azime, Tadesse Destaw Belay, Mesay Gemeda Yigezu, Moges Ahmed Ah Mehamed, Abinew Ali Ayele, Ebrahim Chekol Jibril, Michael Melese Woldeyohannis, Olga Kolesnikova, Philipp Slusallek, Dietrich Klakow, Seid Muhie Yimam

Attention on Multiword Expressions: A Multilingual Study of BERT-based Models with Regard to Idiomaticity and Microsyntax (2025.findings-naacl)

Challenge:	Low-resource languages are lagging behind current state-of-the-art (SOTA) developments in the field of NLP due to insufficient resources to train LLMs.
Approach:	They propose to use multilingual large language models for five Ethiopian languages and a benchmark dataset to evaluate their performance.
Outcome:	The proposed models outperform existing models in five Ethiopian languages and a benchmark dataset for various downstream NLP tasks.

Copied to clipboard

Iuliia Zaitova, Vitalii Hirak, Badr M. Abdullah, Dietrich Klakow, Bernd Möbius, Tania Avgustinova

Syntactic Text Processing

Challenge:	Specifically, models fine-tuned on semantic tasks tend to distribute attention to idiomatic expressions more evenly across layers.
Approach:	They analyze attention patterns of encoder-only models towards two distinct types of Multiword Expressions (MWEs) idioms present challenges in semantic non-compositionality, while MSUs demonstrate unconventional syntactic behavior that does not conform to standard grammatical categorizations.
Outcome:	The proposed models show that fine-tuned models allocate attention to idiomatic expressions more evenly across layers.

The Impact of Demonstrations on Multilingual In-Context Learning: A Multidimensional Analysis (2024.findings-acl)

Copied to clipboard

Miaoran Zhang, Vagrant Gautam, Mingyang Wang, Jesujoba Alabi, Xiaoyu Shen, Dietrich Klakow, Marius Mosbach

A Survey on Recent Approaches for Natural Language Processing in Low-Resource Scenarios (2021.naacl-main)

Challenge:	In-context learning is a popular inference strategy where large language models solve a task using only a few labeled demonstrations without updating the model parameters.
Approach:	They conduct multidimensional analysis of multilingual in-context learning using 5 models from different model families and 9 datasets covering classification and generation tasks.
Outcome:	The results show that demonstrations vary significantly across models, tasks, and languages.

Copied to clipboard

Michael A. Hedderich, Lukas Lange, Heike Adel, Jannik Strötgen, Dietrich Klakow

The Metalogue Debate Trainee Corpus: Data Collection and Annotations (L18-1)

Challenge:	a growing body of work is focused on improving performance in low-resource settings . a goal of this study is to explain how these methods differ in their requirements .
Approach:	They propose to analyze data-lean scenarios across different dimensions of data availability to understand which approaches are effective in a specific low-resource setting.
Outcome:	The proposed methods enable learning when training data is sparse.

Copied to clipboard

Volha Petukhova, Andrei Malchanau, Youssef Oualil, Dietrich Klakow, Saturnino Luz, Fasih Haider, Nick Campbell, Dimitris Koryzis, Dimitris Spiliotopoulos, Pierre Albert, Nicklas Linz, Jan Alexandersson

Challenge:	Argumentation is an important component of human intelligence and is used to train lawyers and citizens in legal domains.
Approach:	They describe the Metalogue Debate Trainee Corpus (DTC) which contains data on motion and speech capture devices and semantic annotations.
Outcome:	The metalogue Debate Trainee Corpus (DTC) was developed to facilitate the design of instructional and interactive models for the Virtual Debate Coach application.

Placing M-Phasis on the Plurality of Hate: A Feature-Based Corpus of Hate Online (2022.lrec-1)

Copied to clipboard

Dana Ruiter, Liane Reiners, Ashwin Geet D’Sa, Thomas Kleinbauer, Dominique Fohr, Irina Illina, Dietrich Klakow, Christian Schemer, Angeliki Monnier

Ethical NLP

Challenge:	HS-related corpora over-simplify the phenomenon of hate by labelling user content with binary classes, e.g., hate/neutral . this ignores the complex and subjective nature of HS, which limits the real-life applicability of classifiers trained on these corporales.
Approach:	They present a corpus of 9k German and french user comments from migration-related news articles.
Outcome:	The proposed corpus is annotated with 23 features that become descriptors of various types of speech, ranging from critical comments to implicit and explicit expressions of hate.

Charting the Landscape of African NLP: Mapping Progress and Shaping the Road Ahead (2025.emnlp-main)

Copied to clipboard

Jesujoba Oluwadara Alabi, Michael A. Hedderich, David Ifeoluwa Adelani, Dietrich Klakow

Challenge:	African languages are often left behind in state-of-the-art natural language processing systems and large language models.
Approach:	They analyze 884 research papers on NLP for African languages published over past five years . they identify key trends shaping the field and outline promising directions .
Outcome:	The findings identify key trends shaping the field and outline promising directions . the authors analyze 884 research papers on NLP for African languages published over the past five years .

Fine-Tuning Large Language Models to Translate: Will a Touch of Noisy Data in Misaligned Languages Suffice? (2024.emnlp-main)

Copied to clipboard

Dawei Zhu, Pinzhen Chen, Miaoran Zhang, Barry Haddow, Xiaoyu Shen, Dietrich Klakow

Cross-lingual Transfer Learning for Japanese Named Entity Recognition (N19-2)

Challenge:	Traditionally, success in multilingual machine translation depends on large volume, diverse directions, and high quality of training data.
Approach:	They revisit the importance of large language models for translation by fine-tuning on 32 parallel sentences.
Outcome:	The proposed model can be fine-tuned on as few as 32 parallel sentences . however, the choice of direction is critical to avoid misinterpretation, the authors say .

Copied to clipboard

Andrew Johnson, Penny Karanasou, Judith Gaspers, Dietrich Klakow

On the Interplay Between Fine-tuning and Sentence-level Probing for Linguistic Knowledge in Pre-trained Transformers (2020.findings-emnlp)

Challenge:	a recent study focuses on bootstrapping named entity models from English to Japanese . TL is a technique that overcomes linguistic differences between the target and source languages .
Approach:	They propose to use a deep neural network model to transfer weights between languages . they also propose a novel approach that romanizes a portion of the Japanese input .
Outcome:	The proposed approach overcomes linguistic differences by romanizing a portion of the Japanese input.

Copied to clipboard

Marius Mosbach, Anna Khokhlova, Michael A. Hedderich, Dietrich Klakow

Evaluating the Capabilities of Large Language Models for Multi-label Emotion Understanding (2025.coling-main)

Challenge:	linguistic knowledge encoded in pre-trained contextual embeddings is poorly understood . fine-tuning can be used to investigate the representations of pre-train models .
Approach:	They propose to investigate fine-tuning of contextualized embedding models through sentence-level probing.
Outcome:	The proposed method improves probing accuracy for three pre-trained models.

Copied to clipboard

Tadesse Destaw Belay, Israel Abebe Azime, Abinew Ali Ayele, Grigori Sidorov, Dietrich Klakow, Philip Slusallek, Olga Kolesnikova, Seid Muhie Yimam

MCSE: Multimodal Contrastive Learning of Sentence Embeddings (2022.naacl-main)

Challenge:	Emotion classification is one of the most challenging tasks in large language models.
Approach:	They propose to use a multi-label emotion classification dataset for four Ethiopian languages to evaluate their ability to learn and reason.
Outcome:	The proposed model improves the understanding of emotions in language models and how people convey emotions through various languages.

Copied to clipboard

Miaoran Zhang, Marius Mosbach, David Adelani, Michael Hedderich, Dietrich Klakow

Who Did You Blame When Your Project Failed? Designing a Corpus for Presupposition Generation in Cross-Examination Dialogues (2024.lrec-main)

Challenge:	Existing approaches to learning semantically meaningful sentence embeddings are limited by the complexity of pre-trained models.
Approach:	They propose a sentence embedding learning approach that exploits both visual and textual information via a multimodal contrastive objective.
Outcome:	The proposed approach improves the state-of-the-art average Spearman’s correlation by 1.7% on a variety of semantic textual similarity tasks.

Copied to clipboard

Maria Francis, Julius Steuer, Dietrich Klakow, Volha Petukhova

Information Retrieval

Challenge:	Existing models for presupposition generation fail to generate complete lists of presuffpositions.
Approach:	They propose to fine-tune existing BERT and T5 models for a task where a model produces a list of presuppositions carried by the given input sentence.
Outcome:	The proposed models outperform BERT and T5 models on the novel task of presupposition as natural language inference (PNLI) despite limited data, they achieved an emerging proficiency in generation of presumptions reaching ROUGE scores of 43.47, adhering to systematic patterns that mirror valid strategies for pres upposition generation, although failed to generate the complete lists.

Annotating Customer-Oriented Behaviour in Call Centre Sales Dialogues (2024.lrec-main)

Copied to clipboard

Jutta Stock, Volha Petukhova, Dietrich Klakow

Natural Language Interfaces

Challenge:	Customer-oriented behaviour (COB) is often hindered by a lack of clarity in its definition and lack of robust analytical, categorization, and computational approaches.
Approach:	They propose a conceptual and empirical framework for customer-oriented behaviour in call centre interactions . they aim to identify facets of COB that positively impact on Customer Satisfaction .
Outcome:	The proposed framework improves our understanding of the dynamics shaping sales strategies in call centres and holds promise for practical applications in optimising customer-agent interactions.

Handling Noisy Labels for Robustly Learning from Self-Training Data for Low-Resource Sequence Labeling (N19-3)

Copied to clipboard

Debjit Paul, Mittul Singh, Michael A. Hedderich, Dietrich Klakow

It’s Not a Walk in the Park! Challenges of Idiom Translation in Speech-to-text Systems (2025.acl-long)

Challenge:	In low-resource environments, self-training is less effective due to unreliable annotations . we combine self-teaching with noise handling to clean the self-labeled data .
Approach:	They propose to combine self-training with noise handling to clean unlabeled data . they propose to model clean and noisy labels separately to improve performance .
Outcome:	The proposed method performs better than baseline methods on Chunking and NER.

Copied to clipboard

Iuliia Zaitova, Badr M. Abdullah, Wei Xue, Dietrich Klakow, Bernd Möbius, Tania Avgustinova

Understanding “Democratization” in NLP and ML Research (2024.emnlp-main)

Challenge:	idioms are defined as words with a figurative meaning not deducible from their individual components.
Approach:	They compare idiom translation as compared to conventional news translation in two languages . they compare MT and SLT systems with MT, Large Language Models and cascaded alternatives .
Outcome:	The proposed systems show better handling of idioms than standard news translation systems.

Copied to clipboard

Arjun Subramonian, Vagrant Gautam, Dietrich Klakow, Zeerak Talat

Linguistic Theories

Challenge:	a large number of NLP and ML papers mention terms related to democracy . authors find that democratization is most frequently used to convey (ease of) access to or use of technologies without meaningfully engaging with theories of democratisation.
Approach:	They analyze papers using the term "democra*" to clarify how it is understood in NLP and ML . they find that democratization is most frequently used to convey (ease of) access to or use of technologies .
Outcome:	The authors analyze papers using the term "democra*" they find that democratization is most frequently used to convey (ease of) access to or use of technologies without meaningfully engaging with theories of democratisation.

MasakhaPOS: Part-of-Speech Tagging for Typologically Diverse African languages (2023.acl-long)

Copied to clipboard

Cheikh M. Bamba Dione, David Ifeoluwa Adelani, Peter Nabende, Jesujoba Alabi, Thapelo Sindane, Happy Buzaaba, Shamsuddeen Hassan Muhammad, Chris Chinenye Emezue, Perez Ogayo, Anuoluwapo Aremu, Catherine Gitau, Derguene Mbaye, Jonathan Mukiibi, Blessing Sibanda, Bonaventure F. P. Dossou, Andiswa Bukula, Rooweither Mabuya, Allahsera Auguste Tapo, Edwin Munkoh-Buabeng, Victoire Memdjokam Koagne, Fatoumata Ouoba Kabore, Amelia Taylor, Godson Kalipe, Tebogo Macucwa, Vukosi Marivate, Tajuddeen Gwadabe, Mboning Tchiaze Elvis, Ikechukwu Onyenwe, Gratien Atindogbe, Tolulope Adelani, Idris Akinade, Olanrewaju Samuel, Marien Nahimana, Théogène Musabeyezu, Emile Niyomutabazi, Ester Chimhenga, Kudzai Gotosa, Patrick Mizha, Apelete Agbolo, Seydou Traore, Chinedu Uchechukwu, Aliyu Yusuf, Muhammad Abdullahi, Dietrich Klakow

Syntactic Text Processing

Challenge:	POS tagging is one of the fundamental steps for many natural language processing (NLP) applications.
Approach:	They present AfricaPOS, the largest part-of-speech (POS) dataset for 20 typologically diverse African languages.
Outcome:	The proposed model improves POS tagging performance in unseen languages.

FAME: Feature-Based Adversarial Meta-Embeddings for Robust Input Representations (2021.emnlp-main)

Copied to clipboard

Lukas Lange, Heike Adel, Jannik Strötgen, Dietrich Klakow

Representation Learning

Challenge:	Recent work on word embeddings and pre-trained language models has shown the large impact of language representations on natural language processing (NLP) models across tasks and domains.
Approach:	They propose feature-based adversarial meta-embeddings with an attention function that is guided by word-specific properties, such as shape and frequency, to handle subword-based embeddings.
Outcome:	The proposed model improves performance in downstream tasks even with word embeddings from transformers.

Familiar words but strange voices: Modelling the influence of speech variability on word recognition (2021.eacl-srw)

Copied to clipboard

Alexandra Mayn, Badr M. Abdullah, Dietrich Klakow

Multimodality

Challenge:	Despite the lack of acoustic-phonetic invariance in speech, listeners can reliably recognize spoken words despite the lack aural-phonemic invariancy.
Approach:	They propose a deep neural model which is trained to retrieve the meaning of a word given its spoken form, a task which resembles that faced by a human listener.
Outcome:	The proposed model is more sensitive to dialectical variation than gender variation and more related to related languages.

ATC-ANNO: Semantic Annotation for Air Traffic Control with Assistive Auto-Annotation (2020.lrec-1)

Copied to clipboard

Marc Schulder, Johannah O’Mahony, Yury Bakanouski, Dietrich Klakow

Speech Recognition

Challenge:	ATC communications are a challenging domain for automatic speech recognition (ASR) due to the time-sensitive nature of their task, annotators must have prior experience with ATC communication.
Approach:	They propose a tool for the transcription and semantic annotation of air traffic communications.
Outcome:	The proposed tool can annotate four times as many utterances in a single time.

Transfer Learning and Distant Supervision for Multilingual Transformer Models: A Study on African Languages (2020.emnlp-main)

Copied to clipboard

Michael A. Hedderich, David Adelani, Dawei Zhu, Jesujoba Alabi, Udia Markus, Dietrich Klakow

Feature-Dependent Confusion Matrices for Low-Resource NER Labeling with Noisy Labels (D19-1)

Challenge:	Recent studies show that results from high-resource languages cannot be easily transferred to realistic, low-resourced scenarios.
Approach:	They analyse performance of multilingual transformer models using available resources for Hausa, isiXhosa and NER and topic classification.
Outcome:	The proposed models can achieve with as little as 10 or 100 labeled sentences the same performance as baselines with much more supervised training data.

Copied to clipboard

Lukas Lange, Michael A. Hedderich, Dietrich Klakow

Information Extraction & Text Mining

Challenge:	Existing approaches to improve supervised labeling with noisy training data do not take the input features into account or they need to learn the noise modeling from scratch.
Approach:	They propose to cluster training data using input features and compute different confusion matrices for each cluster.
Outcome:	The proposed model improves on low-resource named entity recognition settings in several languages, compared with other models which do not take the input features into account or need to learn noise modeling from scratch.

A Closer Look at Linguistic Knowledge in Masked Language Models: The Case of Relative Clauses in American English (2020.coling-main)

Copied to clipboard

Marius Mosbach, Stefania Degaetano-Ortlieb, Marie-Pauline Krielke, Badr M. Abdullah, Dietrich Klakow

MasakhaNER 2.0: Africa-centric Transfer Learning for Named Entity Recognition (2022.emnlp-main)

Challenge:	Despite the high performance of transformer-based language models, we still lack understanding of the kind of linguistic knowledge they learn and rely on.
Approach:	They evaluate three transformer-based language models and test their grammatical and semantic knowledge by sentence-level probing, diagnostic cases, and masked prediction tasks.
Outcome:	The models capture grammatical and semantic knowledge, but they lack model-specific weaknesses especially on semantic knowledge.

Copied to clipboard

David Adelani, Graham Neubig, Sebastian Ruder, Shruti Rijhwani, Michael Beukman, Chester Palen-Michel, Constantine Lignos, Jesujoba Alabi, Shamsuddeen Muhammad, Peter Nabende, Cheikh M. Bamba Dione, Andiswa Bukula, Rooweither Mabuya, Bonaventure F. P. Dossou, Blessing Sibanda, Happy Buzaaba, Jonathan Mukiibi, Godson Kalipe, Derguene Mbaye, Amelia Taylor, Fatoumata Kabore, Chris Chinenye Emezue, Anuoluwapo Aremu, Perez Ogayo, Catherine Gitau, Edwin Munkoh-Buabeng, Victoire Memdjokam Koagne, Allahsera Auguste Tapo, Tebogo Macucwa, Vukosi Marivate, Mboning Tchiaze Elvis, Tajuddeen Gwadabe, Tosin Adewumi, Orevaoghene Ahia, Joyce Nakatumba-Nabende, Neo Lerato Mokono, Ignatius Ezeani, Chiamaka Chukwuneke, Mofetoluwa Oluwaseun Adeyemi, Gilles Quentin Hacheme, Idris Abdulmumin, Odunayo Ogundepo, Oreen Yousuf, Tatiana Moteu, Dietrich Klakow

Emoji-Based Transfer Learning for Sentiment Tasks (2021.eacl-srw)

Challenge:	Existing studies on named entity recognition methods for African languages focus on English as the source language, but there is evidence that it is not the best for low-resource languages.
Approach:	They propose to use human-annotated datasets to analyze named entity recognition tasks in 20 African languages to test whether they are effective.
Outcome:	The proposed method improves zero-shot F1 scores by 14% over 20 languages compared to using English .

Copied to clipboard

Susann Boy, Dana Ruiter, Dietrich Klakow

Multimodality

Challenge:	Sentiment tasks such as hate speech detection and sentiment analysis are often low-resource . a transfer learning approach is used to transfer the emotional information encoded in emojis to a sentiment task .
Approach:	They exploit emotional information encoded in emojis to enhance performance on sentiment tasks . they use a transfer learning approach where parameters learned by an e-based source task are transferred to a sentiment target task .
Outcome:	The proposed method improves sentiment tasks on languages other than English with high emoji content and label distribution under three conditions.

Few-shot Fine-tuning vs. In-context Learning: A Fair Comparison and Evaluation (2023.findings-acl)

Copied to clipboard

Marius Mosbach, Tiago Pimentel, Shauli Ravfogel, Dietrich Klakow, Yanai Elazar

ProverbEval: Exploring LLM Evaluation Challenges for Low-resource Language Understanding (2025.findings-naacl)

Challenge:	Recent studies show that in-context learning and few-shot fine-tuning can generalize well out-of-domain.
Approach:	They compare few-shot fine-tuning and in-context learning for task adaptation . they find that both approaches generalize similarly, but exhibit large variation .
Outcome:	The proposed methods outperform in-context learning and few-shot fine-tuning with OPT models of different sizes.

Copied to clipboard

Israel Abebe Azime, Atnafu Lambebo Tonja, Tadesse Destaw Belay, Yonas Chanie, Bontu Fufa Balcha, Negasi Haile Abadi, Henok Biadglign Ademtew, Mulubrhan Abebe Nerea, Debela Desalegn Yadeta, Derartu Dagne Geremew, Assefa Atsbiha Tesfu, Philipp Slusallek, Thamar Solorio, Dietrich Klakow

The Hidden Space of Transformer Language Adapters (2024.acl-long)

Challenge:	Large language models (LLMs) evaluation is gaining increasing attention as they are typically trained on general-domain datasets while demonstrating notable performance on tasks out of their training domains.
Approach:	They propose an LLM evaluation benchmark for low-resource languages that focuses on low-rsource language understanding in culture-specific scenarios.
Outcome:	The proposed benchmarks outperform monolingual evaluations on proverb generation tasks and native language proverb descriptions on multiple choice tasks.

Copied to clipboard

Jesujoba Alabi, Marius Mosbach, Matan Eyal, Dietrich Klakow, Mor Geva

Modeling the Impact of Syntactic Distance and Surprisal on Cross-Slavic Text Comprehension (2022.lrec-1)

Challenge:	Adapters are small modules trained on top of a frozen language model to adapt predictions to new target languages.
Approach:	They propose to train transformer language adapters on top of a frozen model to adapt predictions to new target languages.
Outcome:	The transformer language adapters are trained on top of a frozen model to adapt predictions to new target languages.

Copied to clipboard

Irina Stenger, Philip Georgis, Tania Avgustinova, Bernd Möbius, Dietrich Klakow

Syntactic Text Processing

Challenge:	Using symmetric measures of insertion, deletion and movement of syntactic units, we investigate phonetic and orthographic asymmetries between selected languages.
Approach:	They focus on the syntactic variation and measure syntaktic distances between nine Slavic languages using symmetric measures of insertion, deletion and movement of syntak units in parallel sentences of the fable “The North Wind and the Sun”.
Outcome:	The proposed measures are validated on spoken and written cloze tests for Slavic native speakers to determine whether variations in syntax lead to slower or impeded intercomprehension of Slav texts.

INJONGO: A Multicultural Intent Detection and Slot-filling Dataset for 16 African Languages (2025.acl-long)

Copied to clipboard

Hao Yu, Jesujoba Oluwadara Alabi, Andiswa Bukula, Jian Yun Zhuang, En-Shiun Annie Lee, Tadesse Kebede Guge, Israel Abebe Azime, Happy Buzaaba, Blessing Kudzaishe Sibanda, Godson Koffi Kalipe, Jonathan Mukiibi, Salomon Kabongo Kabenamualu, Mmasibidi Setaka, Lolwethu Ndolela, Nkiruka Odu, Rooweither Mabuya, Shamsuddeen Hassan Muhammad, Salomey Osei, Sokhar Samb, Dietrich Klakow, David Ifeoluwa Adelani

Bridging the Culture Gap: A Framework for LLM-Driven Socio-Cultural Localization of Math Word Problems in Low-Resource Languages (2026.findings-acl)

Challenge:	Slot-filling and intent detection tasks are well-established tasks in Conversational AI, but current benchmarks for these tasks rely on evaluations of low-resource languages and translations from English benchmarks.
Approach:	They propose to use a multilingual, open-source benchmark dataset for 16 African languages with utterances generated by native speakers across diverse domains.
Outcome:	The proposed dataset compares multilingual transformer models and prompting large language models (LLMs) with the English language.

Copied to clipboard

Israel Abebe Azime, Tadesse Destaw Belay, Dietrich Klakow, Philipp Slusallek, Anshuman Chhabra

A Few Thousand Translations Go a Long Way! Leveraging Pre-trained Models for African News Translation (2022.naacl-main)

Challenge:	Existing multilingual benchmarks that use translations retain English-centric entities.
Approach:	They propose a framework that culturally localizes translated datasets into variants enriched with local entities.
Outcome:	The proposed framework mitigates English-centric entity bias and improves model robustness when native entities are introduced across languages.

Copied to clipboard

David Adelani, Jesujoba Alabi, Angela Fan, Julia Kreutzer, Xiaoyu Shen, Machel Reid, Dana Ruiter, Dietrich Klakow, Peter Nabende, Ernie Chang, Tajuddeen Gwadabe, Freshia Sackey, Bonaventure F. P. Dossou, Chris Emezue, Colin Leong, Michael Beukman, Shamsuddeen Muhammad, Guyo Jarso, Oreen Yousuf, Andre Niyongabo Rubungo, Gilles Hacheme, Eric Peter Wairagala, Muhammad Umair Nasir, Benjamin Ajibade, Tunde Ajayi, Yvonne Gitau, Jade Abbott, Mohamed Ahmed, Millicent Ochieng, Anuoluwapo Aremu, Perez Ogayo, Jonathan Mukiibi, Fatoumata Ouoba Kabore, Godson Kalipe, Derguene Mbaye, Allahsera Auguste Tapo, Victoire Memdjokam Koagne, Edwin Munkoh-Buabeng, Valencia Wagner, Idris Abdulmumin, Ayodele Awokoya, Happy Buzaaba, Blessing Sibanda, Andiswa Bukula, Sam Manthalu

Do we read what we hear? Modeling orthographic influences on spoken word recognition (2021.eacl-srw)

Challenge:	Low-resource languages are left out of large-scale pretraining datasets . authors explore how to leverage existing pre-trained models to create low-resourced translation systems for 16 African languages.
Approach:	They investigate how large-scale pre-trained models can be used to create low-resource translation systems for 16 African languages.
Outcome:	The proposed models can translate between hundreds of languages even though there is little parallel data available for training.

Copied to clipboard

Nicole Macher, Badr M. Abdullah, Harm Brouwer, Dietrich Klakow

Phonology

Challenge:	Existing theories and models of spoken word recognition focus on accessing lexical knowledge given an acoustic realization of a word form.
Approach:	They propose two models that instantiate hypotheses regarding the influence of orthography on spoken word recognition.
Outcome:	The proposed models reproduce human-like behavior in different ways and provide testable hypotheses for future research on the source of orthographic effects in spoken word recognition.

Multilingual Normalization of Temporal Expressions with Masked Language Models (2023.eacl-main)

Copied to clipboard

Lukas Lange, Jannik Strötgen, Heike Adel, Dietrich Klakow