5 papers · 1 filter
Hierarchical Text Classification with LLM-Refined Taxonomies
Jonas Golde, Nicolaas Jedema, Ravi Krishnan +1
Hierarchical text classification (HTC) depends on taxonomies that organize labels into structured hierarchies. However, many real-world taxonomies introduce ambiguities, such as id…
Tuning-Free Personalized Alignment via Trial-Error-Explain In-Context Learning
Hyundong Cho, Karishma Sharma, Nicolaas Jedema +4
Language models are aligned to the collective voice of many, resulting in generic outputs that do not align with specific users' styles. In this work, we present Trial-Error-Explai…
Familiarity: Better Evaluation of Zero-Shot Named Entity Recognition by Quantifying Label Shifts in Synthetic Training Data
Jonas Golde, Patrick Haller, Max Ploner +3
Zero-shot named entity recognition (NER) is the task of detecting named entities of specific types (such as 'Person' or 'Medicine') without any training examples. Current research…
Measuring Retrieval Complexity in Question Answering Systems
Matteo Gabburo, Nicolaas Paul Jedema, Siddhant Garg +2
In this paper, we investigate which questions are challenging for retrieval-based Question Answering (QA). We (i) propose retrieval complexity (RC), a novel metric conditioned on t…
DP-KB: Data Programming with Knowledge Bases Improves Transformer Fine Tuning for Answer Sentence Selection
Nic Jedema, Thuy Vu, Manish Gupta +1
While transformers demonstrate impressive performance on many knowledge intensive (KI) tasks, their ability to serve as implicit knowledge bases (KBs) remains limited, as shown on…