68 citations · 359 across the 29 of their papers we have counts for
32 papers
SENSESHIFT: Continuous Sentiment-Controlled Text Generation via Encoder-based Mask Infilling
Shahed Masoudian, Markus Frohmann, Emmanouil Karystinaios +2
Recent controllable text generation (CTG) for sentiment control has largely focused on decoder-based large language models, making causal attention the dominant paradigm. While eff…
When Deep Research Agents Stagnate: Enhancing Reasoning with Retrieval-Aware Agent Control
Heydar Soudani, Elizabeth Lingg, Faegheh Hasibi +1
In this paper, we analyze the reasoning trajectories of a variety of DRAs and show that existing agents often suffer from reasoning stagnation: the majority of iterations contribut…
Batched Self-Consistency Improves LLM Relevance Assessment and Ranking
Anton Korikov, Pan Du, Scott Sanner +1
LLM query-passage relevance assessment is typically studied using a one-by-one pointwise (PW) strategy where each LLM call judges one passage at a time. However, this strategy requ…
Unlabeled Debiasing in Downstream Tasks via Class-wise Low Variance Regularization
Shahed Masoudian, Markus Frohmann, Navid Rekabsaz +1
Language models frequently inherit societal biases from their training data. Numerous techniques have been proposed to mitigate these biases during both the pre-training and fine-t…
What the Weight?! A Unified Framework for Zero-Shot Knowledge Composition
Carolin Holtermann, Markus Frohmann, Navid Rekabsaz +1
The knowledge encapsulated in a model is the core factor determining its final performance on downstream tasks. Much research in NLP has focused on efficient methods for storing an…
ScaLearn: Simple and Highly Parameter-Efficient Task Transfer by Learning to Scale
Markus Frohmann, Carolin Holtermann, Shahed Masoudian +2
Multi-task learning (MTL) has shown considerable practical benefits, particularly when using language models (LMs). While this is commonly achieved by learning tasks under a jo…
Domain Information Control at Inference Time for Acoustic Scene Classification
Shahed Masoudian, Khaled Koutini, Markus Schedl +2
Domain shift is considered a challenge in machine learning as it causes significant degradation of model performance. In the Acoustic Scene Classification task (ASC), domain shift…
Leveraging Domain Knowledge for Inclusive and Bias-aware Humanitarian Response Entry Classification
Nicolò Tamagnone, Selim Fekih, Ximena Contla +2
Accurate and rapid situation analysis during humanitarian crises is critical to delivering humanitarian aid efficiently and is fundamental to humanitarian imperatives and the Leave…
Enhancing the Ranking Context of Dense Retrieval Methods through Reciprocal Nearest Neighbors
George Zerveas, Navid Rekabsaz, Carsten Eickhoff
Sparse annotation poses persistent challenges to training dense retrieval models; for example, it distorts the training signal when unlabeled relevant documents are used spuriously…
Parameter-efficient Modularised Bias Mitigation via AdapterFusion
Deepak Kumar, Oleg Lesota, George Zerveas +4
Large pre-trained language models contain societal biases and carry along these biases to downstream tasks. Current in-processing bias mitigation approaches (like adversarial train…
Leveraging Vision-Language Models for Granular Market Change Prediction
Christopher Wimmer, Navid Rekabsaz
Predicting future direction of stock markets using the historical data has been a fundamental component in financial forecasting. This historical data contains the information of a…
HumSet: Dataset of Multilingual Information Extraction and Classification for Humanitarian Crisis Response
Selim Fekih, Nicolò Tamagnone, Benjamin Minixhofer +4
Timely and effective response to humanitarian crises requires quick and accurate analysis of large amounts of text data - a process that can highly benefit from expert-assisted NLP…
Unlearning Protected User Attributes in Recommendations with Adversarial Training
Christian Ganhör, David Penz, Navid Rekabsaz +2
Collaborative filtering algorithms capture underlying consumption patterns, including the ones specific to particular demographics or protected information of users, e.g. gender, r…
Modular and On-demand Bias Mitigation with Attribute-Removal Subnetworks
Lukas Hauzenberger, Shahed Masoudian, Deepak Kumar +2
Societal biases are reflected in large pre-trained language models and their fine-tuned versions on downstream tasks. Common in-processing bias mitigation approaches, such as adver…
Do Perceived Gender Biases in Retrieval Results Affect Relevance Judgements?
Klara Krieg, Emilia Parada-Cabaleiro, Markus Schedl +1
This work investigates the effect of gender-stereotypical biases in the content of retrieved results on the relevance judgement of users/annotators. In particular, since relevance…
Grep-BiasIR: A Dataset for Investigating Gender Representation-Bias in Information Retrieval Results
Klara Krieg, Emilia Parada-Cabaleiro, Gertraud Medicus +3
The provided contents by information retrieval (IR) systems can reflect the existing societal biases and stereotypes. Such biases in retrieval results can lead to further establish…
CODER: An efficient framework for improving retrieval through COntextual Document Embedding Reranking
George Zerveas, Navid Rekabsaz, Daniel Cohen +1
Contrastive learning has been the dominant approach to training dense retrieval models. In this work, we investigate the impact of ranking context - an often overlooked aspect of l…
WECHSEL: Effective initialization of subword embeddings for cross-lingual transfer of monolingual language models
Benjamin Minixhofer, Fabian Paischer, Navid Rekabsaz
Large pretrained language models (LMs) have become the central building block of many NLP applications. Training these models requires ever more computational resources and most of…
Analyzing Item Popularity Bias of Music Recommender Systems: Are Different Genders Equally Affected?
Oleg Lesota, Alessandro B. Melchiorre, Navid Rekabsaz +4
Several studies have identified discrepancies between the popularity of items in user profiles and the corresponding recommendation lists. Such behavior, which concerns a variety o…
A Modern Perspective on Query Likelihood with Deep Generative Retrieval Models
Oleg Lesota, Navid Rekabsaz, Daniel Cohen +3
Existing neural ranking models follow the text matching paradigm, where document-to-query relevance is estimated through predicting the matching score. Drawing from the rich litera…
Societal Biases in Retrieved Contents: Measurement Framework and Adversarial Mitigation for BERT Rankers
Navid Rekabsaz, Simone Kopeinik, Markus Schedl
Societal biases resonate in the retrieved contents of information retrieval (IR) systems, resulting in reinforcing existing stereotypes. Approaching this issue requires established…
Not All Relevance Scores are Equal: Efficient Uncertainty and Calibration Modeling for Deep Retrieval Models
Daniel Cohen, Bhaskar Mitra, Oleg Lesota +2
In any ranking system, the retrieval model outputs a single score for a document based on its belief on how relevant it is to a given search query. While retrieval models have cont…
TripClick: The Log Files of a Large Health Web Search Engine
Navid Rekabsaz, Oleg Lesota, Markus Schedl +2
Click logs are valuable resources for a variety of information retrieval (IR) tasks. This includes query understanding/analysis, as well as learning effective IR models particularl…
Do Neural Ranking Models Intensify Gender Bias?
Navid Rekabsaz, Markus Schedl
Concerns regarding the footprint of societal biases in information retrieval (IR) systems have been raised in several previous studies. In this work, we examine various recent IR m…