publications

2026

1. Linear Script Representations in Speech Foundation Models Enable Zero-Shot Transliteration

   Ryan Soh-Eun Shim, Kwanghee Choi, Kalvin Chang, Ming-Hao Hsu, Florian Eichin, and 6 more authors

   In Findings of the Association for Computational Linguistics: ACL 2026, Jul 2026

   Multilingual speech foundation models such as Whisper are trained on web-scale data, where data for each language consists of a myriad of regional varieties. However, different regional varieties often employ different scripts to write the same language, rendering speech recognition output also subject to non-determinism in the output script. To mitigate this problem, we show that script is linearly encoded in the activation space of multilingual speech models, and that modifying activations at inference time enables direct control over output script. We find the addition of such script vectors to activations at test time can induce script change even in unconventional language-script pairings (e.g. Italian in Cyrillic and Japanese in Latin script). We apply this approach to inducing post-hoc control over the script of speech recognition output, where we observe competitive performance across all model sizes of Whisper.

2. Copy First, Translate Later: Interpreting Translation Dynamics in Multilingual Pretraining

   Felicia Körner, Maria Matveev, Florian Eichin, Gitta Kutyniok, Barbara Plank, and 1 more author

   In arxiv 2604.17633, Jul 2026
3. Beyond Input Understanding: Diagnosing Multilingual Mathematical Reasoning with Directed Acyclic Trace Graphs

   Jiaqiao Zhang, Zhoujun Li, Raoyuan Zhao, Jian Lan, Thomas Seidl, and 3 more authors

   In arxiv 2605.27715, Jul 2026
4. AfriqueLLM: How Data Mixing and Model Architecture Impact Continued Pre-training for African Languages

   Hao Yu, Tianyi Xu, Michael A. Hedderich, Wassim Hamidouche, Syed Waqas Zamir, and 1 more author

   In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2026

   Large language models (LLMs) are increasingly multilingual, yet open models continue to underperform relative to proprietary systems, with the gap most pronounced for African languages. Continued pre-training (CPT) offers a practical route to language adaptation, but improvements on demanding capabilities such as mathematical reasoning often remain limited. This limitation is driven in part by the uneven domain coverage and missing task-relevant knowledge that characterize many low-resource language corpora. We present AfriqueLLM, a suite of open LLMs adapted to 20 African languages through CPT on 26B tokens. We perform a comprehensive empirical study across five base models spanning sizes and architectures, including Llama 3.1, Gemma 3, and Qwen 3, and systematically analyze how CPT data composition shapes downstream performance. In particular, we vary mixtures that include math, code, and synthetic translated data, and evaluate the resulting models on a range of multilingual benchmarks. Our results identify data composition as the primary driver of CPT gains. Adding math, code, and synthetic translated data yields consistent improvements, including on reasoning-oriented evaluations. Within a fixed architecture, larger models typically improve performance, but architectural choices dominate scale when comparing across model families. Moreover, strong multilingual performance in the base model does not reliably predict post-CPT outcomes; robust architectures coupled with task-aligned data provide a more dependable recipe. Finally, our best models improve long-context performance, including document-level translation.

5. ReverseMath: Answer Inversion for Scalable and Verifiable Mathematical Problem Generation

   Raoyuan Zhao, Yihong Liu, Yupei Du, Hinrich Schütze, and Michael A. Hedderich

   In arxiv 2605.27709, Jul 2026
6. Crosslingual On-Policy Self-Distillation for Multilingual Reasoning

   Yihong Liu, Raoyuan Zhao, Michael A. Hedderich, and Hinrich Schütze

   In arxiv 2605.09548, Jul 2026
7. Evaluating Robustness of Large Language Models Against Multilingual Typographical Errors

   Raoyuan Zhao, Yihong Liu, Lena Altinger, Hinrich Schuetze, and Michael A. Hedderich

   In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2026

   Large language models (LLMs) are increasingly deployed in multilingual, real-world applications with user inputs – naturally introducing \textittypographical errors (typos). Yet most benchmarks assume clean input, leaving the robustness of LLMs to typos across languages largely underexplored. To address this gap, we introduce MulTypo, a multilingual typo generation algorithm that simulates human-like errors based on language-specific keyboard layouts and typing behavior. We evaluate 18 open-source LLMs across three model families and five downstream tasks spanning language inference, multi-choice question answering, mathematical reasoning, and machine translation tasks. Our results show that typos consistently degrade performance, particularly in generative tasks and those requiring reasoning – while the natural language inference task is comparatively more robust. Instruction tuning improves clean-input performance but may increase brittleness under noise. We also observe language-dependent robustness: high-resource languages are generally more robust than low-resource ones, and translation from English is more robust than translation into English. Our findings underscore the need for noise-aware training and multilingual robustness evaluation. We release a Python package for MulTypo and make the source code publicly available at \urlhttps://github.com/cisnlp/multypo.

8. Large Reasoning Models Are (Not Yet) Multilingual Latent Reasoners

   Yihong Liu, Raoyuan Zhao, Hinrich Schuetze, and Michael A. Hedderich

   In Findings of the Association for Computational Linguistics: ACL 2026, Jul 2026

   Large reasoning models (LRMs) achieve strong performance on mathematical reasoning tasks, often attributed to their capability to generate explicit chain-of-thought (CoT) explanations. However, recent work shows that LRMs often arrive at the correct answer before completing these textual reasoning steps, indicating the presence of latent reasoning – internal, non-verbal computation encoded in hidden states. While this phenomenon has been explored in English, its multilingual behavior remains largely unknown. In this paper, we conduct a systematic investigation of multilingual latent reasoning in LRMs across 11 languages. Using a truncation-based strategy, we examine how the correct answer emerges as the model is given only partial reasoning traces, allowing us to measure stepwise latent prediction formation. Our results reveal clear evidence of multilingual latent reasoning, though unevenly: strong in resource-rich languages, weaker in low-resource ones, and broadly less observable on harder benchmarks. To understand whether these differences reflect distinct internal mechanisms, we further perform representational analyses. Despite surface-level disparities, we find that the internal evolution of predictions is highly consistent across languages and broadly aligns with English – a pattern suggesting an English-centered latent reasoning pathway.

9. ToMigo: Interpretable Design Concept Graphs for Aligning Generative AI with Creative Intent

   Lena Hegemann, Xinyi Wen, Michael A. Hedderich, Tarmo Nurmi, and Hariharan Subramonyam

   In Proceedings of the 2026 Designing Interactive Systems Conference, Jul 2026

   Generative AI often produces results misaligned with user intentions, for example, resolving ambiguous prompts in unexpected ways. Despite existing approaches to clarify intent, a major challenge remains: understanding and influencing AI’s interpretation of user intent through simple, direct inputs requiring no expertise or rigid procedures. We present ToMigo, representing intent as design concept graphs: nodes represent choices of purpose, content, or style, while edges link them with interpretable explanations. Applied to graphic design, ToMigo infers intent from reference images and text. We derived a schema of node types and edges from pre-study data, informing a multimodal large language model to generate graphs aligning nodes externally with user intent and internally toward a unified design goal. This structure enables users to explore AI reasoning and directly manipulate the design concept. In our user studies, ToMigo’s design concept graphs received high alignment ratings and captured most user intentions well. Users reported greater control and found interactive features—editable graphs, reflective chats, concept-design realignment—useful for evolving and realizing their design ideas.

10. From Weights to Activations: Is Steering the Next Frontier of Adaptation?

    Simon Ostermann, Daniil Gurgurov, Tanja Baeumel, Michael A. Hedderich, Sebastian Lapuschkin, and 2 more authors

    In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Jul 2026

    Post-training adaptation of large language models is commonly achieved through parameter updates or input based methods such as fine-tuning, parameter-efficient adaptation, and prompting. In parallel, a growing body of work modifies internal activations at inference time to influence model behavior, an approach known as *steering*. Despite increasing use, steering is rarely analyzed within the same conceptual framework as established adaptation methods.In this work, we argue that steering should be regarded as a form of model adaptation. We introduce a set of functional criteria for adaptation methods and use them to compare steering approaches with classical alternatives. This analysis positions steering as a distinct adaptation paradigm based on targeted interventions in activation space, enabling local and reversible behavioral change without parameter updates. The resulting framing clarifies how steering relates to existing methods, motivating a unified taxonomy for model adaptation.

2025

1. Charting the Landscape of African NLP: Mapping Progress and Shaping the Road Ahead

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

   In EMNLP’25, Jul 2025
2. What’s the Difference? Supporting Users in Identifying the Effects of Prompt and Model Changes Through Token Patterns

   Michael A. Hedderich, Anyi Wang, Raoyuan Zhao, Florian Eichin, Jonas Fischer, and 1 more author

   In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL), Jul 2025

   Prompt engineering for large language models is challenging, as even small prompt perturbations or model changes can significantly impact the generated output texts. Existing evaluation methods of LLM outputs, either automated metrics or human evaluation, have limitations, such as providing limited insights or being labor-intensive. We propose Spotlight, a new approach that combines both automation and human analysis. Based on data mining techniques, we automatically distinguish between random (decoding) variations and systematic differences in language model outputs. This process provides token patterns that describe the systematic differences and guide the user in manually analyzing the effects of their prompts and changes in models efficiently. We create three benchmarks to quantitatively test the reliability of token pattern extraction methods and demonstrate that our approach provides new insights into established prompt data. From a human-centric perspective, through demonstration studies and a user study, we show that our token pattern approach helps users understand the systematic differences of language model outputs. We are further able to discover relevant differences caused by prompt and model changes (e.g. related to gender or culture), thus supporting the prompt engineering process and human-centric model behavior research.

3. Probing LLMs for Multilingual Discourse Generalization Through a Unified Label Set

   Florian Eichin, Yang Janet Liu, Barbara Plank, and Michael A. Hedderich

   In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (ACL), Jul 2025

   Discourse understanding is essential for many NLP tasks, yet most existing work remains constrained by framework-dependent discourse representations. This work investigates whether large language models (LLMs) capture discourse knowledge that generalizes across languages and frameworks. We address this question along two dimensions: (1) developing a unified discourse relation label set to facilitate cross-lingual and cross-framework discourse analysis, and (2) probing LLMs to assess whether they encode generalizable discourse abstractions. Using multilingual discourse relation classification as a testbed, we examine a comprehensive set of 23 LLMs of varying sizes and multilingual capabilities. Our results show that LLMs, especially those with multilingual training corpora, can generalize discourse information across languages and frameworks. Further layer-wise analyses reveal that language generalization at the discourse level is most salient in the intermediate layers. Lastly, our error analysis provides an account of challenging relation classes.

4. Explaining crowdworker behaviour through computational rationality

   Michael A. Hedderich, and Antti Oulasvirta

   In Behaviour & Information Technology, Jul 2025
5. Grokking ExPLAIND: Unifying Model, Data, and Training Attribution to Study Model Behavior

   Florian Eichin, Yupei Du, Philipp Mondorf, Barbara Plank, and Michael A. Hedderich

   In Accepted at ICML’26, Jul 2025
6. MAKIEval: A Multilingual Automatic WiKidata-based Framework for Cultural Awareness Evaluation for LLMs

   Raoyuan Zhao, Beiduo Chen, Barbara Plank, and Michael A. Hedderich

   In Findings of EMNLP’25, Jul 2025
7. Do We Know What LLMs Don’t Know? A Study of Consistency in Knowledge Probing

   Raoyuan Zhao, Abdullatif Köksal, Ali Modarressi, Michael A. Hedderich, and Hinrich Schütze

   In Findings of EMNLP’25, Jul 2025
8. Semantic Component Analysis: Discovering Patterns in Short Texts Beyond Topics

   Florian Eichin, Carolin Schuster, Georg Groh, and Michael A. Hedderich

   In Accepted at Findings of EMNLP’25, Jul 2025

2024

1. A Piece of Theatre: Investigating How Teachers Design LLM Chatbots to Assist Adolescent Cyberbullying Education

   Michael A. Hedderich, Natalie N. Bazarova, Wenting Zou, Ryun Shim, Xinda Ma, and 1 more author

   Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems, Jul 2024
2. The Potential and Challenges of Evaluating Attitudes, Opinions, and Values in Large Language Models

   Bolei Ma, Xinpeng Wang, Tiancheng Hu, Anna-Carolina Haensch, Michael A. Hedderich, and 2 more authors

   In EMNLP Findings, Jul 2024
3. Understanding When Tree of Thoughts Succeeds: Larger Models Excel in Generation, Not Discrimination

   Qiqi Chen, Xinpeng Wang, Philipp Mondorf, Michael A. Hedderich, and Barbara Plank

   In Arxiv Preprint, Jul 2024
4. Understanding Human-AI Workflows for Generating Personas

   Joongi Shin, Michael A. Hedderich, Bartłomiej Jakub Rey, Andrés Lucero, and Antti Oulasvirta

   In Proceedings of the 2024 ACM Designing Interactive Systems Conference (DIS), Jul 2024
5. Facilitating Asynchronous Idea Generation and Selection with Chatbots

   Joongi Shin, Ankit Khatri, Michael A. Hedderich, Andrés Lucero, and Antti Oulasvirta

   In Proceedings of OzCHI 2024, Jul 2024

2023

1. An exploration of knowledge‐organizing technologies to advance transdisciplinary back pain research

   Jeffrey C. Lotz, Glen Ropella, Paul Anderson, Qian Yang, Michael A. Hedderich, and 2 more authors

   In JOR SPINE, Jul 2023

   Abstract Chronic low back pain (LBP) is influenced by a broad spectrum of patient-specific factors as codified in domains of the biopsychosocial model (BSM). Operationalizing the BSM into research and clinical care is challenging because most investigators work in silos that concentrate on only one or two BSM domains. Furthermore, the expanding, multidisciplinary nature of BSM research creates practical limitations as to how individual investigators integrate current data into their processes of generating impactful hypotheses. The rapidly advancing field of artificial intelligence (AI) is providing new tools for organizing knowledge, but the practical aspects for how AI may advance LBP research and clinical are beginning to be explored. The goals of the work presented here are to: (1) explore the current capabilities of knowledge integration technologies (large language models (LLM), similarity graphs (SGs), and knowledge graphs (KGs)) to synthesize biomedical literature and depict multimodal relationships reflected in the BSM, and; (2) highlight limitations, implementation details, and future areas of research to improve performance. We demonstrate preliminary evidence that LLMs, like GPT-3, may be useful in helping scientists analyze and distinguish cLBP publications across multiple BSM domains and determine the degree to which the literature supports or contradicts emergent hypotheses. We show that SG representations and KGs enable exploring LBP’s literature in novel ways, possibly providing, trans-disciplinary perspectives or insights that are currently difficult, if not infeasible to achieve. The SG approach is automated, simple, and inexpensive to execute, and thereby may be useful for early-phase literature and narrative explorations beyond one’s areas of expertise. Likewise, we show that KGs can be constructed using automated pipelines, queried to provide semantic information, and analyzed to explore trans-domain linkages. The examples presented support the feasibility for LBP-tailored AI protocols to organize knowledge and support developing and refining trans-domain hypotheses.

2. Meta Self-Refinement for Robust Learning with Weak Supervision

   Dawei Zhu, Xiaoyu Shen, Michael A. Hedderich, and Dietrich Klakow

   In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics (EACL), Jul 2023

   Training deep neural networks (DNNs) under weak supervision has attracted increasing research attention as it can significantly reduce the annotation cost. However, labels from weak supervision can be noisy, and the high capacity of DNNs enables them to easily overfit the label noise, resulting in poor generalization. Recent methods leverage self-training to build noise-resistant models, in which a teacher trained under weak supervision is used to provide highly confident labels for teaching the students. Nevertheless, the teacher derived from such frameworks may have fitted a substantial amount of noise and therefore produce incorrect pseudo-labels with high confidence, leading to severe error propagation. In this work, we propose Meta Self-Refinement (MSR), a noise-resistant learning framework, to effectively combat label noise from weak supervision. Instead of relying on a fixed teacher trained with noisy labels, we encourage the teacher to refine its pseudo-labels. At each training step, MSR performs a meta gradient descent on the current mini-batch to maximize the student performance on a clean validation set. Extensive experimentation on eight NLP benchmarks demonstrates that MSR is robust against label noise in all settings and outperforms state-of-the-art methods by up to 11.4% in accuracy and 9.26% in F1 score.

3. SparseIMU: Computational Design of Sparse IMU Layouts for Sensing Fine-Grained Finger Microgestures

   Adwait Sharma, Christina Salchow-Hömmen, Vimal Suresh Mollyn, Aditya Shekhar Nittala, Michael A. Hedderich, and 3 more authors

   ACM Trans. Comput.-Hum. Interact., Jul 2023

   Gestural interaction with freehands and while grasping an everyday object enables always-available input. To sense such gestures, minimal instrumentation of the user’s hand is desirable. However, the choice of an effective but minimal IMU layout remains challenging, due to the complexity of the multi-factorial space that comprises diverse finger gestures, objects and grasps. We present SparseIMU, a rapid method for selecting minimal inertial sensor-based layouts for effective gesture recognition. Furthermore, we contribute a computational tool to guide designers with optimal sensor placement. Our approach builds on an extensive microgestures dataset that we collected with a dense network of 17 inertial measurement units (IMUs). We performed a series of analyses, including an evaluation of the entire combinatorial space for freehand and grasping microgestures (393K layouts), and quantified the performance across different layout choices, revealing new gesture detection opportunities with IMUs. Finally, we demonstrate the versatility of our method with four scenarios.

4. Understanding and Mitigating Classification Errors Through Interpretable Token Patterns

   Michael A. Hedderich, Jonas Fischer, Dietrich Klakow, and Jilles Vreeken

   In Abstract at BlackboxNLP @EMNLP’23, Jul 2023
5. From an Analog to a Digital Workflow: An Introductory Approach to Digital Editions in Assyriology

   Timo Homburg, Tim Brandes, Eva-Maria Huber, and Michael A. Hedderich

   Cuneiform Digital Library Bulletin, Jul 2023

2022

1. Chatbots Facilitating Consensus-Building in Asynchronous Co-Design

   Joongi Shin, Michael A. Hedderich, Andrés Lucero, and Antti Oulasvirta

   In Proceedings of the 35th Annual ACM Symposium on User Interface Software and Technology (UIST), Jul 2022

   Consensus-building is an essential process for the success of co-design projects. To build consensus, stakeholders need to discuss conflicting needs and viewpoints, converge their ideas toward shared interests, and grow their willingness to commit to group decisions. However, managing group discussions is challenging in large co-design projects with multiple stakeholders. In this paper, we investigate the interaction design of a chatbot that can mediate consensus-building conversationally. By interacting with individual stakeholders, the chatbot collects ideas to satisfy conflicting needs and engages stakeholders to consider others’ viewpoints, without having stakeholders directly interact with each other. Results from an empirical study in an educational setting (N = 12) suggest that the approach can increase stakeholders’ commitment to group decisions and maintain the effect even on the group decisions that conflict with personal interests. We conclude that chatbots can facilitate consensus-building in small-to-medium-sized projects, but more work is needed to scale up to larger projects.

2. MCSE: Multimodal Contrastive Learning of Sentence Embeddings

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

   In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Jul 2022

   Learning semantically meaningful sentence embeddings is an open problem in natural language processing. In this work, we propose a sentence embedding learning approach that exploits both visual and textual information via a multimodal contrastive objective. Through experiments on a variety of semantic textual similarity tasks, we demonstrate that our approach consistently improves the performance across various datasets and pre-trained encoders. In particular, combining a small amount of multimodal data with a large text-only corpus, we improve the state-of-the-art average Spearman’s correlation by 1.7%. By analyzing the properties of the textual embedding space, we show that our model excels in aligning semantically similar sentences, providing an explanation for its improved performance.

3. Weak Supervision and Label Noise Handling for Natural Language Processing in Low-Resource Scenarios

   Michael A. Hedderich

   Jul 2022
4. Label-Descriptive Patterns and Their Application to Characterizing Classification Errors

   Michael A. Hedderich, Jonas Fischer, Dietrich Klakow, and Jilles Vreeken

   In International Conference on Machine Learning (ICML), Jul 2022
5. Task-Adaptive Pre-Training for Boosting Learning With Noisy Labels: A Study on Text Classification for African Languages

   Dawei Zhu, Michael A. Hedderich, Fangzhou Zhai, David Ifeoluwa Adelani, and Dietrich Klakow

   In Proceedings of the ICLR 2022 Workshop AfricaNLP, Jul 2022
6. Is BERT Robust to Label Noise? A Study on Learning with Noisy Labels in Text Classification

   Dawei Zhu, Michael A. Hedderich, Fangzhou Zhai, David Adelani, and Dietrich Klakow

   In Proceedings of the Third Workshop on Insights from Negative Results in NLP, Jul 2022

   Incorrect labels in training data occur when human annotators make mistakes or when the data is generated via weak or distant supervision. It has been shown that complex noise-handling techniques - by modeling, cleaning or filtering the noisy instances - are required to prevent models from fitting this label noise. However, we show in this work that, for text classification tasks with modern NLP models like BERT, over a variety of noise types, existing noise-handling methods do not always improve its performance, and may even deteriorate it, suggesting the need for further investigation. We also back our observations with a comprehensive analysis.

2021

1. On the Correlation of Context-Aware Language Models With the Intelligibility of Polish Target Words to Czech Readers

   Klára Jágrová, Michael Hedderich, Marius Mosbach, Tania Avgustinova, and Dietrich Klakow

   Frontiers in Psychology, Jul 2021

   This contribution seeks to provide a rational probabilistic explanation for the intelligibility of words in a genetically related language that is unknown to the reader, a phenomenon referred to as intercomprehension. In this research domain, linguistic distance, among other factors, was proved to correlate well with the mutual intelligibility of individual words. However, the role of context for the intelligibility of target words in sentences was subject to very few studies. To address this, we analyze data from web-based experiments in which Czech (CS) respondents were asked to translate highly predictable target words at the final position of Polish sentences. We compare correlations of target word intelligibility with data from 3-g language models (LMs) to their correlations with data obtained from context-aware LMs. More specifically, we evaluate two context-aware LM architectures: Long Short-Term Memory (LSTMs) that can, theoretically, take infinitely long-distance dependencies into account and Transformer-based LMs which can access the whole input sequence at the same time. We investigate how their use of context affects surprisal and its correlation with intelligibility.

2. Proceedings of the First Workshop on Weakly Supervised Learning (WeaSuL)

   Jul 2021
3. Proceedings of the Third Workshop on Beyond Vision and LANguage: inTEgrating Real-world kNowledge (LANTERN)

   Jul 2021
4. Estimating Formulas for Model Performance Under Noisy Labels Using Symbolic Regression

   In Proceedings of the 31th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN), Jul 2021
5. A Survey on Recent Approaches for Natural Language Processing in Low-Resource Scenarios

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

   In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT), Jul 2021
6. Analysing the Noise Model Error for Realistic Noisy Label Data

   Michael A. Hedderich, Dawei Zhu, and Dietrich Klakow

   In Thirty-Fifth AAAI Conference on Artificial Intelligence, Jul 2021
7. SoloFinger: Robust Microgestures While Grasping Everyday Objects

   Adwait Sharma, Michael A. Hedderich, Divyanshu Bhardwaj, Bruno Fruchard, Jess McIntosh, and 4 more authors

   In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, Jul 2021

   Using microgestures, prior work has successfully enabled gestural interactions while holding objects. Yet, these existing methods are prone to false activations caused by natural finger movements while holding or manipulating the object. We address this issue with SoloFinger, a novel concept that allows design of microgestures that are robust against movements that naturally occur during primary activities. Using a data-driven approach, we establish that single-finger movements are rare in everyday hand-object actions and infer a single-finger input technique resilient to false activation. We demonstrate this concept’s robustness using a white-box classifier on a pre-existing dataset comprising 36 everyday hand-object actions. Our findings validate that simple SoloFinger gestures can relieve the need for complex finger configurations or delimiting gestures and that SoloFinger is applicable to diverse hand-object actions. Finally, we demonstrate SoloFinger’s high performance on commodity hardware using random forest classifiers.

8. ANEA: Distant Supervision for Low-Resource Named Entity Recognition

   Michael A. Hedderich, Lukas Lange, and Dietrich Klakow

   In ICML 2021 Workshop on Practical Machine Learning For Developing Countries, Jul 2021
9. Rekonstruktion von fragmentierten Dokumenten mit NLP

   Johannes Bernhard, and Michael A. Hedderich

   In IDCS Workshop Von Analog zu Digital, Jul 2021

2020

1. Transfer Learning and Distant Supervision for Multilingual Transformer Models: A Study on African Languages

   Michael A. Hedderich, David Adelani, Dawei Zhu, Jesujoba Alabi, Udia Markus, and 1 more author

   In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), Jul 2020

   Multilingual transformer models like mBERT and XLM-RoBERTa have obtained great improvements for many NLP tasks on a variety of languages. However, recent works also showed that results from high-resource languages could not be easily transferred to realistic, low-resource scenarios. In this work, we study trends in performance for different amounts of available resources for the three African languages Hausa, isiXhosa and on both NER and topic classification. We show that in combination with transfer learning or distant supervision, these models can achieve with as little as 10 or 100 labeled sentences the same performance as baselines with much more supervised training data. However, we also find settings where this does not hold. Our discussions and additional experiments on assumptions such as time and hardware restrictions highlight challenges and opportunities in low-resource learning.

2. On the Interplay Between Fine-tuning and Sentence-Level Probing for Linguistic Knowledge in Pre-Trained Transformers

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

   In Proceedings of the Third BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP, Nov 2020

   Fine-tuning pre-trained contextualized embedding models has become an integral part of the NLP pipeline. At the same time, probing has emerged as a way to investigate the linguistic knowledge captured by pre-trained models. Very little is, however, understood about how fine-tuning affects the representations of pre-trained models and thereby the linguistic knowledge they encode. This paper contributes towards closing this gap. We study three different pre-trained models: BERT, RoBERTa, and ALBERT, and investigate through sentence-level probing how fine-tuning affects their representations. We find that for some probing tasks fine-tuning leads to substantial changes in accuracy, possibly suggesting that fine-tuning introduces or even removes linguistic knowledge from a pre-trained model. These changes, however, vary greatly across different models, fine-tuning and probing tasks. Our analysis reveals that while fine-tuning indeed changes the representations of a pre-trained model and these changes are typically larger for higher layers, only in very few cases, fine-tuning has a positive effect on probing accuracy that is larger than just using the pre-trained model with a strong pooling method. Based on our findings, we argue that both positive and negative effects of fine-tuning on probing require a careful interpretation.

3. Distant Supervision and Noisy Label Learning for Low Resource Named Entity Recognition: A Study on Hausa and Yorùbá

   David Ifeoluwa Adelani, Michael A. Hedderich, Dawei Zhu, Esther Berg, and Dietrich Klakow

   Nov 2020

2019

1. Using Multi-Sense Vector Embeddings for Reverse Dictionaries

   Michael A. Hedderich, Andrew Yates, Dietrich Klakow, and Gerard Melo

   In Proceedings of the 13th International Conference on Computational Semantics - Long Papers, May 2019

   Popular word embedding methods such as word2vec and GloVe assign a single vector representation to each word, even if a word has multiple distinct meanings. Multi-sense embeddings instead provide different vectors for each sense of a word. However, they typically cannot serve as a drop-in replacement for conventional single-sense embeddings, because the correct sense vector needs to be selected for each word. In this work, we study the effect of multi-sense embeddings on the task of reverse dictionaries. We propose a technique to easily integrate them into an existing neural network architecture using an attention mechanism. Our experiments demonstrate that large improvements can be obtained when employing multi-sense embeddings both in the input sequence as well as for the target representation. An analysis of the sense distributions and of the learned attention is provided as well.

2. Handling Noisy Labels for Robustly Learning from Self-Training Data for Low-Resource Sequence Labeling

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

   In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Student Research Workshop, Jun 2019

   In this paper, we address the problem of effectively self-training neural networks in a low-resource setting. Self-training is frequently used to automatically increase the amount of training data. However, in a low-resource scenario, it is less effective due to unreliable annotations created using self-labeling of unlabeled data. We propose to combine self-training with noise handling on the self-labeled data. Directly estimating noise on the combined clean training set and self-labeled data can lead to corruption of the clean data and hence, performs worse. Thus, we propose the Clean and Noisy Label Neural Network which trains on clean and noisy self-labeled data simultaneously by explicitly modelling clean and noisy labels separately. In our experiments on Chunking and NER, this approach performs more robustly than the baselines. Complementary to this explicit approach, noise can also be handled implicitly with the help of an auxiliary learning task. To such a complementary approach, our method is more beneficial than other baseline methods and together provides the best performance overall.

3. Feature-Dependent Confusion Matrices for Low-Resource NER Labeling with Noisy Labels

   Lukas Lange, Michael A. Hedderich, and Dietrich Klakow

   In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Jun 2019

   In low-resource settings, the performance of supervised labeling models can be improved with automatically annotated or distantly supervised data, which is cheap to create but often noisy. Previous works have shown that significant improvements can be reached by injecting information about the confusion between clean and noisy labels in this additional training data into the classifier training. However, for noise estimation, these approaches either do not take the input features (in our case word embeddings) into account, or they need to learn the noise modeling from scratch which can be difficult in a low-resource setting. We propose to cluster the training data using the input features and then compute different confusion matrices for each cluster. To the best of our knowledge, our approach is the first to leverage feature-dependent noise modeling with pre-initialized confusion matrices. We evaluate on low-resource named entity recognition settings in several languages, showing that our methods improve upon other confusion-matrix based methods by up to 9%.

2018

1. Training a Neural Network in a Low-Resource Setting on Automatically Annotated Noisy Data

   Michael A. Hedderich, and Dietrich Klakow

   In Proceedings of the Workshop on Deep Learning Approaches for Low-Resource NLP, Jun 2018

   Manually labeled corpora are expensive to create and often not available for low-resource languages or domains. Automatic labeling approaches are an alternative way to obtain labeled data in a quicker and cheaper way. However, these labels often contain more errors which can deteriorate a classifier’s performance when trained on this data. We propose a noise layer that is added to a neural network architecture. This allows modeling the noise and train on a combination of clean and noisy data. We show that in a low-resource NER task we can improve performance by up to 35% by using additional, noisy data and handling the noise.