Wolfgang Nejdl

TMLR Journal 2026 Journal Article

Robust Conformal Prediction for Infrequent Classes

• Jens-Michalis Papaioannou
• Sebastian Jäger
• Alexei Figueroa
• David Stutz
• Betty van Aken
• Keno Bressem
• Wolfgang Nejdl
• Felix Gers

Many real-world classification tasks involve datasets with large and imbalanced label spaces, making class-specific uncertainty quantification particularly challenging. Conformal Prediction (CP) provides a model-agnostic framework, which formally guarantees coverage, meaning that its prediction sets contain the true label with a user-defined probability (confidence level). However, standard class-conditional methods often fail when data is scarce for some classes. We propose a method that uses domain knowledge or label hierarchies to dynamically group semantically related classes to meet the desired coverage for a given confidence threshold. Our method maintains class-conditioned calibration when possible and provides group-conditioned guarantees where necessary. We evaluate our method on outcome diagnoses prediction, an important clinical task that does not only benefit from robust uncertainty estimation, but also presents a very imbalanced label distribution. We conduct experiments using three clinical datasets employing two medical taxonomies (ICD-10 and CCSR) and label spaces of varying sizes with up to more than 1,000 classes. Our results show that the proposed approach is able to successfully exploit the label hierarchy and consistently improves class-conditional coverage for infrequent diagnoses. By improving coverage for underrepresented classes, our method enhances the reliability and trustworthiness of predictive models. This improvement is especially valuable in clinical applications, where failure to detect rare but serious conditions can lead to harmful consequences.

ECAI Conference 2025 Conference Paper

"Where does it hurt? " - Dataset and Study on Physician Intent Trajectories in Doctor Patient Dialogues

• Tom Röhr
• Soumyadeep Roy
• Fares Al Mohamad
• Jens-Michalis Papaioannou
• Wolfgang Nejdl
• Felix A. Gers
• Alexander Löser

In a doctor-patient dialogue, the primary objective of physicians is to diagnose patients and propose a treatment plan. Medical doctors guide these conversations through targeted questioning to efficiently gather the information required to provide the best possible outcomes for patients. To the best of our knowledge, this is the first work that studies physician intent trajectories in doctor-patient dialogues. We use the ‘Ambient Clinical Intelligence Benchmark’ (Aci-bench) dataset for our study. We collaborate with medical professionals to develop a fine-grained taxonomy of physician intents based on the SOAP framework (Subjective, Objective, Assessment, and Plan). We then conduct a large-scale annotation effort to label over 5000 doctor-patient turns with the help of a large number of medical experts recruited using Prolific, a popular crowd-sourcing platform. This large labeled dataset is an important resource contribution that we use for benchmarking the state-of-the-art generative and encoder models for medical intent classification tasks. Our findings show that our models understand the general structure of medical dialogues with high accuracy, but often fail to identify transitions between SOAP categories. We also report for the first time common trajectories in medical dialogue structures that provide valuable insights for designing ‘differential diagnosis’ systems. Finally, we extensively study the impact of intent filtering for medical dialogue summarization and observe a significant boost in performance. We make the codes and data, including annotation guidelines, publicly available at https: //github. com/DATEXIS/medical-intent-classification.

AAMAS Conference 2025 Conference Paper

Enhancing Offline Reinforcement Learning with Curriculum Learning-Based Trajectory Valuation

• Amir Abolfazli
• Zekun Song
• Avishek Anand
• Wolfgang Nejdl

The success of deep reinforcement learning (DRL) relies on the availability and quality of training data, often requiring extensive interactions with specific environments. In many real-world scenarios, where data collection is costly and risky, offline reinforcement learning (RL) offers a solution by utilizing data collected by domain experts and searching for a batch-constrained optimal policy. This approach is further augmented by incorporating external data sources, expanding the range and diversity of data collection possibilities. However, existing offline RL methods often struggle with challenges posed by non-matching data from these external sources. In this work, we specifically address the problem of source-target domain mismatch in scenarios involving mixed datasets, characterized by a predominance of source data generated from random or suboptimal policies and a limited amount of target data generated from higher-quality policies. To tackle this problem, we introduce Transition Scoring (TS), a novel method that assigns scores to transitions based on their similarity to the target domain, and propose Curriculum Learning-Based Trajectory Valuation (CLTV), which effectively leverages these transition scores to identify and prioritize high-quality trajectories through a curriculum learning approach. Our extensive experiments across various offline RL methods and MuJoCo environments, complemented by rigorous theoretical analysis, demonstrate that CLTV enhances the overall performance and transferability of policies learned by offline RL algorithms.

IJCAI Conference 2025 Conference Paper

Enhancing Online Climate Discourse: A Two-Stage Framework for Climate Content Categorization and Moderation

• Apoorva Upadhyaya
• Wolfgang Nejdl
• Marco Fisichella

Climate change is one of the most pressing global challenges that requires urgent adaptation and resilience efforts, highlighting the need for both scientific solutions and effective communication. In the digital age, online content plays a key role in shaping climate narratives. Therefore, previous research has mainly focused on public perception or categorized content by topics such as impacts, mitigation, policy, etc. Despite these efforts, identifying discussions that address climate change adaptation is crucial for monitoring resilience and assessing public sentiment, while recognizing denial narratives helps combat misinformation. Moreover, the public's exposure to online climate content can either lead to or hinder climate action, emphasizing the need for climate content moderation. To address these issues, we propose a novel multi-stage framework where stage 1 categorizes climate-related content into adaptation, resilience, and denial while stage 2 moderates content by enhancing or intervening based on its alignment with climate goals. We present a novel dataset by manually annotating publicly available tweets and news articles into different climate categories with the help of a taxonomy developed by domain experts. Extensive experiments with benchmark climate and other domain datasets validate the efficacy of our prediction stage, while human and external evaluations confirm the relevance of our moderation stage.

TMLR Journal 2025 Journal Article

SPONGE: Competing Sparse Language Representations for Effective Knowledge Transfer

• Jens-Michalis Papaioannou
• Alexei Figueroa
• Conor Fallon
• Anna Capilla
• Alexandra Bekiaridou
• Stavros Zanos
• Wolfgang Nejdl
• Alexander Löser

In domains with privacy constraints, most knowledge resides in siloed datasets, hindering the development of a model with all relevant knowledge for a task. Clinical NLP is a prime example of these constraints in practice. Research in this area typically falls back to the canonical setting of sequential transfer learning, where a model pre-trained on large corpora is finetuned on a smaller annotated dataset. An avenue for knowledge transfer among diverse clinics is multi-step sequential transfer learning since models are more likely to be shared than private clinical data. This setting poses challenges of cross-linguality, domain diversity, and varying label distributions which undermine generalisation. We propose SPONGE, an efficient prototypical architecture that leverages competing sparse language representations. These encompass distributed knowledge and create the necessary level of redundancy for effective transfer learning across multiple datasets. We identify that prototypical classifiers are critically sensitive to label-recency bias which we mitigate with a novel strategy at inference time. SPONGE in combination with this strategy significantly boosts generalisation performance to unseen data. With the help of medical professionals, we show that the explainability of our models is clinically relevant. We make all source code available.

AAAI Conference 2024 Conference Paper

FairTrade: Achieving Pareto-Optimal Trade-Offs between Balanced Accuracy and Fairness in Federated Learning

• Maryam Badar
• Sandipan Sikdar
• Wolfgang Nejdl
• Marco Fisichella

As Federated Learning (FL) gains prominence in distributed machine learning applications, achieving fairness without compromising predictive performance becomes paramount. The data being gathered from distributed clients in an FL environment often leads to class imbalance. In such scenarios, balanced accuracy rather than accuracy is the true representation of model performance. However, most state-of-the-art fair FL methods report accuracy as the measure of performance, which can lead to misguided interpretations of the model's effectiveness to mitigate discrimination. To the best of our knowledge, this work presents the first attempt towards achieving Pareto-optimal trade-offs between balanced accuracy and fairness in a federated environment (FairTrade). By utilizing multi-objective optimization, the framework negotiates the intricate balance between model's balanced accuracy and fairness. The framework's agnostic design adeptly accommodates both statistical and causal fairness notions, ensuring its adaptability across diverse FL contexts. We provide empirical evidence of our framework's efficacy through extensive experiments on five real-world datasets and comparisons with six baselines. The empirical results underscore the potential of our framework in improving the trade-off between fairness and balanced accuracy in FL applications.

AAAI Conference 2024 Conference Paper

IVP-VAE: Modeling EHR Time Series with Initial Value Problem Solvers

• Jingge Xiao
• Leonie Basso
• Wolfgang Nejdl
• Niloy Ganguly
• Sandipan Sikdar

Continuous-time models such as Neural ODEs and Neural Flows have shown promising results in analyzing irregularly sampled time series frequently encountered in electronic health records. Based on these models, time series are typically processed with a hybrid of an initial value problem (IVP) solver and a recurrent neural network within the variational autoencoder architecture. Sequentially solving IVPs makes such models computationally less efficient. In this paper, we propose to model time series purely with continuous processes whose state evolution can be approximated directly by IVPs. This eliminates the need for recurrent computation and enables multiple states to evolve in parallel. We further fuse the encoder and decoder with one IVP solver utilizing its invertibility, which leads to fewer parameters and faster convergence. Experiments on three real-world datasets show that the proposed method can systematically outperform its predecessors, achieve state-of-the-art results, and have significant advantages in terms of data efficiency.

AAMAS Conference 2024 Conference Paper

Multi-Robot Motion and Task Planning in Automotive Production Using Controller-based Safe Reinforcement Learning

• Eric Wete
• Joel Greenyer
• Daniel Kudenko
• Wolfgang Nejdl

Using synthesis- and AI-planning-based approaches, recent works investigated methods to support engineers with the automation of design, planning, and execution of multi-robot cells. However, real-time constraints and stochastic processes were not well covered due, e. g. , to the high abstraction level of the problem modeling, and these methods do not scale well. In this paper, using probabilistic model checking, we construct a controller and integrate it with reinforcement learning approaches to synthesize the most efficient and correct multi-robot task schedules. Statistical Model Checking (SMC) is applied for system requirement verification. Our method is aware of uncertainties and considers robot movement times, interruption times, and stochastic interruptions that can be learned during multi-robot cell operations. We developed a model-at-runtime that integrates the execution of the production cell and optimizes its performance using a controller-based AI system. For this purpose and to derive the best policy, we implemented and compared AI-based methods, namely, Monte Carlo Tree Search, a heuristic AI-planning technique, and Q-learning, a model-free reinforcement learning method. Our results show that our methodology can choose time-efficient task sequences that consequently improve the cycle time and efficiently adapt to stochastic events, e. g. , robot interruptions. Moreover, our approach scales well compared to previous investigations using SMC, which did not reveal any violation of the requirements.

ECAI Conference 2024 Conference Paper

TrustFed: Navigating Trade-offs Between Performance, Fairness, and Privacy in Federated Learning

• Maryam Badar
• Sandipan Sikdar
• Wolfgang Nejdl
• Marco Fisichella

As Federated Learning (FL) gains prominence in secure machine learning applications, achieving trustworthy predictions without compromising predictive performance becomes paramount. While Differential Privacy (DP) is extensively used for its effective privacy protection, yet its application as a lossy protection method can lower the predictive performance of the machine learning model. Also, the data being gathered from distributed clients in an FL environment often leads to class imbalance making traditional accuracy measure less reflective of the true performance of prediction model. In this context, we introduce a fairness-aware FL framework (TrustFed) based on Gaussian differential privacy and Multi-Objective Optimization (MOO), which effectively protects privacy while providing fair and accurate predictions. To the best of our knowledge, this is the first attempt towards achieving Pareto-optimal trade-offs between balanced accuracy and fairness in a federated environment while safeguarding the privacy of individual clients. The framework’s flexible design adeptly accommodates both statistical parity and equal opportunity fairness notions, ensuring its applicability in various FL scenarios. We demonstrate our framework’s effectiveness through comprehensive experiments on five real-world datasets. TrustFed consistently achieves comparable performance fairness tradeoff to the state-of-the-art (SoTA) baseline models while preserving the anonymization rights of users in FL applications.

ECAI Conference 2023 Conference Paper

GENEMASK: Fast Pretraining of Gene Sequences to Enable Few-Shot Learning

• Soumyadeep Roy
• Jonas Wallat
• Sowmya S. Sundaram
• Wolfgang Nejdl
• Niloy Ganguly

Large-scale language models such as DNABert and LOGO aim to learn optimal gene representations and are trained on the entire Human Reference Genome. However, standard tokenization schemes involve a simple sliding window of tokens like k-mers that do not leverage any gene-based semantics and thus may lead to (trivial) masking of easily predictable sequences, and subsequently inefficient Masked Language Modeling (MLM) training. Therefore, we propose a novel masking algorithm, GENEMASK, for MLM training of gene sequences, where we randomly identify positions in a gene sequence as mask centers and locally select the span around the mask center with the highest Normalized Pointwise Mutual Information (NPMI) to mask. We observe that in the absence of human-understandable semantics in the genomics domain (in contrast, semantic units like words and phrases are inherently available in NLP), GENEMASK-based models substantially outperform the SOTA models (DNABert and LOGO) over four benchmark gene sequence classification datasets in five few-shot settings (10 to 1000-shot). More significantly, the GENEMASK-based DNABert model is trained for less than one-tenth of the number of epochs of the original SOTA model. We also observe a strong correlation between top-ranked PMI tokens and conserved DNA sequence motifs, which may indicate the incorporation of latent genomic information. The codes (including trained models) and datasets are made publicly available at https: //github. com/roysoumya/GeneMask.

JBHI Journal 2023 Journal Article

Guest Editorial Trustworthy and Collaborative AI for Personalised Healthcare Through Edge-of-Things

• Zhao Ren
• Björn W. Schuller
• Björn M. Eskofier
• Thanh Tam Nguyen
• Wolfgang Nejdl

From diagnosis to therapies, the development of artificial intelligence (AI) has facilitated improvements in personalised healthcare applications. The evolution of AI in healthcare is closely related to the changes in the types and volumes of data which we need to deal with. The first generation of healthcare technologies, represented by the highly successful relational databases, are designed to handle structured data involving patient demographics, patient care, treatments, and outcomes of those treatments. Big Data platforms, which are representative of the current mainstream healthcare technologies, are built to process unstructured data from sources like electronic health records, medical imaging, genomic sequencing, and pharmaceutical research. The next generation of healthcare technologies will potentially be Edge-of-Things data, represented by massive amount of streaming data generated from Internet-of-Things frameworks, Cloud systems, and Edge computing platforms.

IJCAI Conference 2023 Conference Paper

Intensity-Valued Emotions Help Stance Detection of Climate Change Twitter Data

• Apoorva Upadhyaya
• Marco Fisichella
• Wolfgang Nejdl

Our study focuses on the United Nations Sustainable Development Goal 13: Climate Action, by identifying public attitudes on Twitter about climate change. Public consent and participation is the key factor in dealing with climate crises. However, discussions about climate change on Twitter are often influenced by the polarised beliefs that shape the discourse and divide it into communities of climate change deniers and believers. In our work, we propose a framework that helps identify different attitudes in tweets about climate change (deny, believe, ambiguous). Previous literature often lacks an efficient architecture or ignores the characteristics of climate-denier tweets. Moreover, the presence of various emotions with different levels of intensity turns out to be relevant for shaping discussions on climate change. Therefore, our paper utilizes emotion recognition and emotion intensity prediction as auxiliary tasks for our main task of stance detection. Our framework injects the words affecting the emotions embedded in the tweet to capture the overall representation of the attitude in terms of the emotions associated with it. The final task-specific and shared feature representations are fused with efficient embedding and attention techniques to detect the correct attitude of the tweet. Extensive experiments on our novel curated dataset, two publicly available climate change datasets (ClimateICWSM-2023 and ClimateStance-2022), and a benchmark dataset for stance detection (SemEval-2016) validate the effectiveness of our approach.

TIST Journal 2013 Journal Article

Introduction to the special section on twitter and microblogging services

• Irwin King
• Wolfgang Nejdl

AIIM Journal 1997 Journal Article

Abstract temporal diagnosis in medical domains

• Johann Gamper
• Wolfgang Nejdl

Most current model-based diagnosis formalisms and algorithms are defined only for static systems, which is often inadequate for medical reasoning. In this paper we describe a model-based framework plus algorithms for diagnosing time-dependent systems where we can define qualitative temporal scenarios. Complex temporal behavior is described within a logical framework extended by qualitative temporal constraints. Abstract observations aggregate from observations at time points to assumptions over time intervals. These concepts provide a very natural representation and make diagnosis independent of the number of actual observations and the temporal resolution. The concept of abstract temporal diagnosis captures in a natural way the kind of indefinite temporal knowledge which is frequently available in medical diagnoses. We use viral hepatitis B (including a set of real hepatitis B data) to illustrate and evaluate our framework. The comparison of our results with the results of Hepaxpert-I is promising. The diagnosis computed in our system is often more precise than the diagnosis in Hepaxpert-I and we detect inconsistent data sequences which cannot be detected in the latter system.

ECAI Conference 1996 Conference Paper

A Model-based Reasoning Approach to Circumscription

• Peter Fröhlich 0001
• Wolfgang Nejdl

IJCAI Conference 1995 Conference Paper

Proposing Measurements in Dynamic Systems

• Johann Gamper
• Wolfgang Nejdl

Proposing measurements in diagnosis systems for static systems is a well understood task. Usually, entropy based approaches are used, sometimes extended by cost and other considerations. How to do the same task in dynamic systems is less clear, and so far measurement proposal algorithms have been ignored in the recent approaches advanced for dynamic systems. In this paper we will describe a set of techniques and algorithms suitable for measurement proposal in a temporal diagnosis formalism discussed in our previous work. This formalism is based on qualitative Allen constraints. The current paper introduces a measurement proposal algorithm and improves it in several ways. Finally an entropy-based computation method is described for this temporal setting.

ECAI Conference 1994 Conference Paper

Harnessing the Power of Temporal Abstractions in Model-Based Diagnosis of Dynamic Systems

• Wolfgang Nejdl
• Johann Gamper

UAI Conference 1994 Conference Paper

Model-based Diagnosis with Qualitative Temporal Uncertainty

• Wolfgang Nejdl
• Johann Gamper

In this paper we describe a framework for model-based diagnosis of dynamic systems, which extends previous work in this field by using and expressing temporal uncertainty in the form of qualitative interval relations a la Allen. Based on a logical framework extended by qualitative and quantitative temporal constraints we show how to describe behavioral models (both consistency- and abductive-based), discuss how to use abstract observations and show how abstract temporal diagnoses are computed. This yields an expressive framework, which allows the representation of complex temporal behavior allowing us to represent temporal uncertainty. Due to its abstraction capabilities computation is made independent of the number of observations and time points in a temporal setting. An example of hepatitis diagnosis is used throughout the paper.

ECAI Conference 1992 Conference Paper

Formaiizing the Repair Process

• Gerhard Friedrich
• Georg Gottlob
• Wolfgang Nejdl

IJCAI Conference 1991 Conference Paper

Integrating Model-Based Monitoring and Diagnosis of Complex Dynamic Systems

• Franz Lackinger
• Wolfgang Nejdl

We present a new approach to model-based monitoring and diagnosis of dynamic systems. The presented DIAMON algorithm1 uses hierarchical models to monitor and diagnose dynamic systems. DIAMON is based on the integration of teleological parameter-based monitoring models and repair-oriented device-based diagnosis models. It combines consistencybased diagnosis with model-based monitoring and uses an extension of the QSIM-language for the representation of qualitative system models. Furthermore, DIAMON is able to detect and localize a broad range of non-permanent faults and thus extends traditional diagnosis which exclusively deals with permanent faulty behavior. The operation of DIAMON will be demonstrated on a real-world example in a multiple-faults scenario.

AAAI Conference 1991 Conference Paper

The P-Systems: A Systematic Classification of Logics of Nonmonotonicity

• Wolfgang Nejdl

In the last years many logics of nonmonotonicity have been developed using various different formalisms and axiomatizations which makes them very difficult to compare. We develop a classification scheme for these logics using only a few simple concepts and axioms based on conditional logics, properties of partial preorders of possible states (worlds) and centering assumptions. Our framework (the P-Systems) allows us to discuss the similarities, main differences and possible extensions of these logics in a simple and natural way.

ECAI Conference 1990 Conference Paper

Redefining the Candidate Space in Model-Based Diagnosis

• Gerhard Friedrich
• Franz Lackinger
• Wolfgang Nejdl