← Search

Jens Kleesiek

9 accepted papers

2025

A Modular Approach for Clinical SLMs Driven by Synthetic Data with Pre-Instruction Tuning, Model Merging, and Clinical-Tasks Alignment

ACL 2025long

High computation costs and latency of large language models such as GPT-4 have limited their deployment in clinical settings. Small language models (SLMs) offer a cost-effective alternative, but their limited capacity requires biomedical domain adaptation, which remains challenging. An additional bo…

Cited by 0SourcePDFScholar
2025

Every Component Counts: Rethinking the Measure of Success for Medical Semantic Segmentation in Multi-Instance Segmentation Tasks

AAAI 2025technical

We present Connected-Component (CC)-Metrics, a novel semantic segmentation evaluation protocol, targeted to align existing semantic segmentation metrics to a multi-instance detection scenario in which each connected component matters. We motivate this setup in the common medical scenario of semantic…

2025

Flatness is Necessary, Neural Collapse is Not: Rethinking Generalization via Grokking

NeurIPS 2025poster

Neural collapse, i.e., the emergence of highly symmetric, class-wise clustered representations, is frequently observed in deep networks and is often assumed to reflect or enable generalization. In parallel, flatness of the loss landscape has been theoretically and empirically linked to generalizatio…

Cited by 0SourceScholar
2025

Little Is Enough: Boosting Privacy by Sharing Only Hard Labels in Federated Semi-Supervised Learning

AAAI 2025technical

In many critical applications, sensitive data is inherently distributed and cannot be centralized due to privacy concerns. A wide range of federated learning approaches have been proposed to train models locally at each client without sharing their sensitive data, typically by exchanging model param…

2025

Towards Conditioning Clinical Text Generation for User Control

ACL 2025finding

Deploying natural language generation systems in clinical settings remains challenging despite advances in Large Language Models (LLMs), which continue to exhibit hallucinations and factual inconsistencies, necessitating human oversight. This paper explores automated dataset augmentation using LLMs…

2024

Comprehensive Study on German Language Models for Clinical and Biomedical Text Understanding

COLING 2024main

Recent advances in natural language processing (NLP) can be largely attributed to the advent of pre-trained language models such as BERT and RoBERTa. While these models demonstrate remarkable performance on general datasets, they can struggle in specialized domains such as medicine, where unique dom…

Cited by 7SourcePDFScholar
2024

Skeleton Recall Loss for Connectivity Conserving and Resource Efficient Segmentation of Thin Tubular Structures

ECCV 2024poster

"Accurately segmenting thin tubular structures, such as vessels, nerves, roads or concrete cracks, is a crucial task in computer vision. Standard deep learning-based segmentation loss functions, such as Dice or Cross-Entropy, focus on volumetric overlap, often at the expense of preserving structural…

2023

On the Impact of Cross-Domain Data on German Language Models

EMNLP 2023long findings

Traditionally, large language models have been either trained on general web crawls or domain-specific data. However, recent successes of generative large language models, have shed light on the benefits of cross-domain datasets. To examine the significance of prioritizing data diversity over qualit…

Cited by 0SourceScholar
2022

Reference-Guided Pseudo-Label Generation for Medical Semantic Segmentation

AAAI 2022technical

Producing densely annotated data is a difficult and tedious task for medical imaging applications. To address this problem, we propose a novel approach to generate supervision for semi-supervised semantic segmentation. We argue that visually similar regions between labeled and unlabeled images lik…

Cited by 75SourcePDFScholar