Papers by Matthias Aßenmacher
GUARD: Glocal Uncertainty-Aware Robust Decoding for Effective and Efficient Open-Ended Text Generation (2025.findings-emnlp)
Copied to clipboard
Yuanhao Ding, Esteban Garces Arias, Meimingwei Li, Julian Rodemann, Matthias Aßenmacher, Danlu Chen, Gaojuan Fan, Christian Heumann, Chongsheng Zhang
| Challenge: | GUARD is a self-adaptive decoding method that balances coherence with diversity in open-ended text generation. |
| Approach: | They propose a self-adaptive decoding method that balances coherence and diversity . they combine global entropy estimates with local entropic deviations to integrate uncertainty . |
| Outcome: | GUARD achieves a good balance between diversity and coherence while exhibiting significant improvements in generation speed. |
Divergent Token Metrics: Measuring degradation to prune away LLM components – and optimize quantization (2024.naacl-long)
Copied to clipboard
Björn Deiseroth, Max Meuer, Nikolas Gritsch, Constantin Eichenberg, Patrick Schramowski, Matthias Aßenmacher, Kristian Kersting
| Challenge: | Large Language Models (LLMs) have reshaped natural language processing with impressive capabilities, but their ever-increasing size has raised concerns about their effective deployment and the need for LLM compression. |
| Approach: | This study introduces the Divergent Token Metrics (DTMs) that measure token divergences that allow deeper insights into the subtleties of model compression. |
| Outcome: | The proposed measures can identify outliers and improve performance in the sparseness of the LLMs. |
taz2024full: Analysing German Newspapers for Gender Bias and Discrimination across Decades (2025.findings-acl)
Copied to clipboard
| Challenge: | a large corpus of German newspaper articles is available for free in other languages, such as English. |
| Approach: | They propose to use taz2024full to analyse gender representation across four decades of reporting. |
| Outcome: | The proposed corpus supports a wide range of applications from diachronic language analysis to critical media studies. |
Algorithmic Fidelity of Large Language Models in Generating Synthetic German Public Opinions: A Case Study (2025.acl-long)
Copied to clipboard
Bolei Ma, Berk Yoztyurk, Anna-Carolina Haensch, Xinpeng Wang, Markus Herklotz, Frauke Kreuter, Barbara Plank, Matthias Aßenmacher
| Challenge: | Recent advances in large language models have generated significant interest in their potential for synthetic data generation across various domains. |
| Approach: | They use open-ended survey data from the German Longitudinal Election Studies to prompt different LLMs to generate synthetic public opinions reflective of German subpopulations by incorporating demographic features into the persona prompts. |
| Outcome: | The LLM performs better for supporters of left-leaning parties like The Greens and The Left compared to other parties, and matches the least with the right-party AfD. |
Can Calibration of Positional Encodings Enhance Long Context Utilization? (2026.findings-eacl)
Copied to clipboard
| Challenge: | Large language models suffer from positional biases that reduce effective utilization of long contexts. |
| Approach: | They propose a training-free framework for calibrating Positional Encodings at inference time. |
| Outcome: | The proposed framework improves on needle-in-a-haystack and cross-chunk reasoning benchmarks and provides a lightweight method for improving long-context utilization. |
Adaptive Contrastive Search: Uncertainty-Guided Decoding for Open-Ended Text Generation (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to decode text to the most probable sequence have been proposed to address these challenges by improving coherence, diversity, and resemblance to human-generated text. |
| Approach: | They propose a novel decoding strategy that extends contrastive search by incorporating an adaptive degeneration penalty informed by the model’s estimated uncertainty at each generation step. |
| Outcome: | The proposed approach improves creativity and coherence while maintaining coherency across model architectures, languages, and datasets. |
Min-k Sampling: Decoupling Truncation from Temperature Scaling via Relative Logit Dynamics (2026.acl-long)
Copied to clipboard
Yuanhao Ding, Meimingwei Li, Esteban Garces Arias, Matthias Aßenmacher, Christian Heumann, Chongsheng Zhang
| Challenge: | Existing methods for decoding large language models have extreme sensitivity to temperature parameter T. |
| Approach: | They propose a dynamic truncation strategy that analyzes the local shape of the logit distribution to identify "semantic cliffs" they show that Min-k consistently improves text quality even under extreme temperature settings . |
| Outcome: | The proposed method achieves strict temperature invariance and low sensitivity to hyperparameter choices. |
promptolution: A Unified, Modular Framework for Prompt Optimization (2026.eacl-demo)
Copied to clipboard
| Challenge: | Existing implementations of prompt optimization are tied to unmaintained, isolated codebases or require invasive integration into application frameworks. |
| Approach: | They propose a unified, modular open-source framework that integrates multiple contemporary discrete prompt optimizers within a single extensible system for both practitioners and researchers. |
| Outcome: | The proposed framework integrates multiple discrete prompt optimizers, supports systematic and reproducible benchmarking, and returns framework-agnostic prompt strings, enabling seamless integration into existing LLM pipelines while remaining agnosite to the underlying model implementation. |
Automatic Transcription of Handwritten Old Occitan Language (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing approaches to handwritten text recognition have shown promising results, but low-resource languages often lack resources. |
| Approach: | They propose an HTR approach that leverages the Transformer architecture for recognizing handwritten Old Occitan language. |
| Outcome: | The proposed approach surpasses state-of-the-art models for Old Occitan HTR, including open-source Transformer-based models and commercial applications like Google Cloud Vision. |
Evaluating Unsupervised Representation Learning for Detecting Stances of Fake News (2020.coling-main)
Copied to clipboard
| Challenge: | Using unsupervised representation learning, automated Fake News detection is a challenge for researchers. |
| Approach: | They examine pre-trained language models with respect to their performance on two Fake News related data sets. |
| Outcome: | The proposed models outperform the autoregression-based models on two Fake News related data sets. |