arXiv subjects

Andreas Fischer

Publications and source records attributed to Andreas Fischer.

At least 19 recordsLinked to original sources

arXiv2026

Reasoning Before Translation: Enhancing Legal Machine Translation with Structured Reasoning

Neural machine translation (NMT) in the legal domain is a linguistically and conceptually demanding task, primarily due to the complexity of legal language and the high level of precision it requires. The recent emergence of reasoning-capable language models opens new possibilities for tackling such challenges. They add to a set of other previously proposed techniques to enhance the translation quality, which includes supervised fine-tuning and reinforcement learning. In this work, we perform a comparison between these various approaches. More particularly, we evaluate small language models such as Qwen3.5 4B, Qwen3.5 9B, and Gemma 3 12B enhanced with various re-training paradigms and compare their performances against frontier reasoning models. We focus on the Swiss legal system, which -- with its unique multilingual statutes -- offers a particularly challenging testbed for reasoning-augmented models. Our results show that the quality of small ``base'' models can be greatly enhanced, and that reinforcement learning with verifiable rewards can be applied to NMT in the legal domain and surpasses the translation quality of supervised fine-tuning. The performance of enhanced small models is close to the one of state-of-the-art reasoning models yet remains inferior. We also note that re-training paradigms yield diminishing returns as model size increase. The code and models are publicly available at https://github.com/aixiuxiuxiu/Legal-MT-SFT-RL.

cs.CL↗

arXiv2026

Evaluating Chunking Strategies for Retrieval-Augmented Generation on Academic Texts

Retrieval-Augmented Generation (RAG) systems use the question-answering capabilities of Large Language Models (LLMs) to access information outside their parameters. We evaluate if cluster-based semantic chunking improves retrieval and answer quality compared to fixed-size and recursive chunking evaluating on long, structured academic theses using the Retrieval Augmented Generation Assessment (RAGAs) framework. RAGAs based faithfulness shows limited reliability in this setup. Performance on fixed versus document specific questions varied substantially, likely related to the formatting of documents and preprocessing. Under the tested configuration, cluster-based chunking did not outperform simpler strategies.

cs.IR↗

arXiv2026

Handwriting Extraction and Analysis of Signature Lists in Swiss Popular Initiatives

Popular initiatives and referendums are central to Swiss democracy, yet the validation of handwritten signature lists remains a labor-intensive manual process. This paper investigates the potential of automated document analysis methods, including OCR and AI-based handwriting analysis, to support this task. We propose a pipeline combining template-based line segmentation with text recognition and writer retrieval techniques, evaluated on a dataset of 443 handwritten entries from 418 writers. Results show that OCR struggles with out-of-vocabulary handwriting, with a CER of 29.6% for first names. In contrast, writer retrieval performs more robustly, reaching an mAP of 50.6%. Furthermore, our experiments indicate that off-the-shelf OCR systems are not sufficiently reliable for transcription of handwritten signature data, particularly for short, out-of-vocabulary entries such as names or addresses. However, writer retrieval methods can effectively identify visually similar entries across signature lists, making them a suitable tool for supporting the detection of potential duplicate submissions based on handwriting similarity.

cs.CV↗

arXiv2026

BullingerDB: A Dataset for Handwritten Text Recognition and Writer Retrieval

We present BullingerDB, a large-scale benchmark dataset for historical document analysis based on the correspondence of Heinrich Bullinger (1504-1575). The corpus comprises 20,898 pages and 499,222 text lines written by 796 writers over six decades, featuring stylistic variation, multilingual content (mostly Latin and Early New High German) as well as meta-information such as writer identity and time. We evaluate BullingerDB on text recognition and writer retrieval. TrOCR, the best performing model, achieves a CER of 9.1%. For writer retrieval, we introduce a temporal nDCG metric to assess time-aware retrieval. While temporally coherent retrieval is achievable, mAP (78.3%) scores indicate challenges due to long-term stylistic variation. With BullingerDB, we aim to establish a new benchmark for multilingual historical text recognition and temporally-aware writer analysis.

cs.CV↗

arXiv2026

From Pen Strokes to Sleep States: Detecting Low-Recovery Days Using Sigma-Lognormal Handwriting Features

While handwriting has traditionally been studied for character recognition and disease classification, its potential to reflect day-to-day physiological fluctuations in healthy individuals remains unexplored. This study examines whether daily variations in sleep-related recovery states can be inferred from online handwriting dynamics. % We propose a personalized binary classification framework that detects low-recovery days using features derived from the Sigma-Lognormal model, which captures the neuromotor generation process of pen strokes. In a 28-day in-the-wild study involving 13 university students, handwriting was recorded three times daily, and nocturnal cardiac indicators were measured using a wearable ring. For each participant, the lowest (or highest) quartile of four sleep-related metrics -- HRV, lowest heart rate, average heart rate, and total sleep duration -- defined the positive class. Leave-One-Day-Out cross-validation showed that PR-AUC significantly exceeded the baseline (0.25) for all four variables after FDR correction, with the strongest performance observed for cardiac-related variables. Importantly, classification performance did not differ significantly across task types or recording timings, indicating that recovery-related signals are embedded in general movement dynamics. These results demonstrate that subtle within-person autonomic recovery fluctuations can be detected from everyday handwriting, opening a new direction for non-invasive, device-independent health monitoring.

cs.HC↗

arXiv2026

Prediction of Grade, Gender, and Academic Performance of Children and Teenagers from Handwriting Using the Sigma-Lognormal Model

Digital handwriting acquisition enables the capture of detailed temporal and kinematic signals reflecting the motor processes underlying writing behavior. While handwriting analysis has been extensively explored in clinical or adult populations, its potential for studying developmental and educational characteristics in children remains less investigated. In this work, we examine whether handwriting dynamics encode information related to student characteristics using a large-scale online dataset collected from Japanese students from elementary school to junior high school. We systematically compare three families of handwriting-derived features: basic statistical descriptors of kinematic signals, entropy-based measures of variability, and parameters obtained from the sigma-lognormal model. Although the dataset contains dense stroke-level recordings, features are aggregated at the student level to enable a controlled comparison between representations. These features are evaluated across three prediction tasks: grade prediction, gender classification, and academic performance classification, using Linear or Logistic Regression and Random Forest models under consistent experimental settings. The results show that handwriting dynamics contain measurable signals related to developmental stage and individual differences, especially for the grade prediction task. These findings highlight the potential of kinematic handwriting analysis and confirm that through their development, children's handwriting evolves toward a lognormal motor organization.

cs.HC↗

arXiv2026

Spatial PDE-aware Selective State-space with Nested Memory for Mobile Traffic Grid Forecasting

Traffic forecasting in cellular networks is a challenging spatiotemporal prediction problem due to strong temporal dependencies, spatial heterogeneity across cells, and the need for scalability to large network deployments. Traditional cell-specific models incur prohibitive training and maintenance costs, while global models often fail to capture heterogeneous spatial dynamics. Recent spatiotemporal architectures based on attention or graph neural networks improve accuracy but introduce high computational overhead, limiting their applicability in large-scale or real-time settings. We study spatiotemporal grid forecasting, where each time step is a 2D lattice of traffic values, and predict the next grid patch using previous patches. We propose NeST-S6, a convolutional selective state-space model (SSM) with a spatial PDE-aware core, implemented in a nested learning paradigm: convolutional local spatial mixing feeds a spatial PDE-aware SSM core, while a nested-learning long-term memory is updated by a learned optimizer when one-step prediction errors indicate unmodeled dynamics. On the mobile-traffic grid (Milan dataset) at three resolutions (202, 502, 1002), NeST-S6 attains lower errors than a strong Mamba-family baseline in both single-step and 6-step autoregressive rollouts. Under drift stress tests, our model's nested memory lowers MAE by 48-65% over a no-memory ablation. NeST-S6 also speeds full-grid reconstruction by 32 times and reduces MACs by 4.3 times compared to competitive per-pixel scanning models, while achieving 61% lower per-pixel RMSE.

cs.LG↗

arXiv2025

Fusion-Based Neural Generalization for Predicting Temperature Fields in Industrial PET Preform Heating

Accurate and efficient temperature prediction is critical for optimizing the preheating process of PET preforms in industrial microwave systems prior to blow molding. We propose a novel deep learning framework for generalized temperature prediction. Unlike traditional models that require extensive retraining for each material or design variation, our method introduces a data-efficient neural architecture that leverages transfer learning and model fusion to generalize across unseen scenarios. By pretraining specialized neural regressor on distinct conditions such as recycled PET heat capacities or varying preform geometries and integrating their representations into a unified global model, we create a system capable of learning shared thermal dynamics across heterogeneous inputs. The architecture incorporates skip connections to enhance stability and prediction accuracy. Our approach reduces the need for large simulation datasets while achieving superior performance compared to models trained from scratch. Experimental validation on two case studies material variability and geometric diversity demonstrates significant improvements in generalization, establishing a scalable ML-based solution for intelligent thermal control in manufacturing environments. Moreover, the approach highlights how data-efficient generalization strategies can extend to other industrial applications involving complex physical modeling with limited data.

cs.LG↗

arXiv2025

Teacher-Student Guided Inverse Modeling for Steel Final Hardness Estimation

Predicting the final hardness of steel after heat treatment is a challenging regression task due to the many-to-one nature of the process -- different combinations of input parameters (such as temperature, duration, and chemical composition) can result in the same hardness value. This ambiguity makes the inverse problem, estimating input parameters from a desired hardness, particularly difficult. In this work, we propose a novel solution using a Teacher-Student learning framework. First, a forward model (Teacher) is trained to predict final hardness from 13 metallurgical input features. Then, a backward model (Student) is trained to infer plausible input configurations from a target hardness value. The Student is optimized by leveraging feedback from the Teacher in an iterative, supervised loop. We evaluate our method on a publicly available tempered steel dataset and compare it against baseline regression and reinforcement learning models. Results show that our Teacher-Student framework not only achieves higher inverse prediction accuracy but also requires significantly less computational time, demonstrating its effectiveness and efficiency for inverse process modeling in materials science.

cs.AI↗

arXiv2025

Evaluating SAP Joule for Code Generation

SAP has released its own proprietary generative model SAP Joule, intended for various generative tasks, including serving as a code assistant for software engineers. While Joule is yet not focused on SAP-specific ABAP code generation, it can be used for other common languages, including Javascript. This paper compares SAP Joules Javascript coding capabilities against a total of 29 other models using the HumanEval-X Javascript benchmark. SAP Joule achieves a strict accuracy of 80.49% as the fifth best model in our evaluation. To the best of our knowledge, this is the first comparative evaluation of SAP Joule code generation capabilities.

cs.SE↗

arXiv2025

Rule-Based Reinforcement Learning for Document Image Classification with Vision Language Models

Rule-based reinforcement learning has been gaining popularity ever since DeepSeek-R1 has demonstrated its success through simple verifiable rewards. In the domain of document analysis, reinforcement learning is not as prevalent, even though many downstream tasks may benefit from the emerging properties of reinforcement learning, particularly the enhanced reason capabilities. We study the effects of rule-based reinforcement learning with the task of Document Image Classification which is one of the most commonly studied downstream tasks in document analysis. We find that reinforcement learning tends to have better generalisation capabilities to out-of-distritbution data, which we examine in three different scenarios, namely out-of-distribution images, unseen classes and different modalities. Our code is available at https://github.com/jungomi/vision-finetune.

cs.CV↗

arXiv2025

A Robust Scheduling of Cyclic Traffic for Integrated Wired and Wireless Time-Sensitive Networks

Time-Sensitive Networking (TSN) is a toolbox of technologies that enable deterministic communication over Ethernet. A key area has been TSN's time-aware traffic shaping (TAS), which supports stringent end-to-end latency and reliability requirements. Configuration of TAS requires the computation of a network-wide traffic schedule, which is particularly challenging with integrated wireless networks (e.g., 5G, Wi-Fi) due to the stochastic nature of wireless links. This paper introduces a novel method for configuring TAS, focusing on cyclic traffic patterns and jitter of wireless links. We formulate a linear program that computes a network-wide time-aware schedule, robust to wireless performance uncertainties. The given method enables robust scheduling of multiple TSN frames per transmission window using a tunable robustness parameter ({\Gamma}). To reduce computational complexity, we also propose a sequential batch-scheduling heuristic that runs in polynomial time. Our approach is evaluated by using different network topologies and wireless link characteristics, demonstrating that the heuristic can schedule 90% of 6500 requested TSN streams in a large topology.

cs.NI↗

arXiv2025

Compressed Learning for Nanosurface Deficiency Recognition Using Angle-resolved Scatterometry Data

Nanoscale manufacturing requires high-precision surface inspection to guarantee the quality of the produced nanostructures. For production environments, angle-resolved scatterometry offers a non- invasive and in-line compatible alternative to traditional surface inspection methods, such as scanning electron microscopy. However, angle-resolved scatterometry currently suffers from long data acquisition time. Our study addresses the issue of slow data acquisition by proposing a compressed learning framework for the accurate recognition of nanosurface deficiencies using angle-resolved scatterometry data. The framework uses the particle swarm optimization algorithm with a sampling scheme customized for scattering patterns. This combination allows the identification of optimal sampling points in scatterometry data that maximize the detection accuracy of five different levels of deficiency in ZnO nanosurfaces. The proposed method significantly reduces the amount of sampled data while maintaining a high accuracy in deficiency detection, even in noisy environments. Notably, by sampling only 1% of the data, the method achieves an accuracy of over 86%, which further improves to 94% when the sampling rate is increased to 6%. These results demonstrate a favorable balance between data reduction and classification performance. The obtained results also show that the compressed learning framework effectively identifies critical sampling areas.

eess.SP↗

arXiv2025

CTC Transcription Alignment of the Bullinger Letters: Automatic Improvement of Annotation Quality

Handwritten text recognition for historical documents remains challenging due to handwriting variability, degraded sources, and limited layout-aware annotations. In this work, we address annotation errors - particularly hyphenation issues - in the Bullinger correspondence, a large 16th-century letter collection. We introduce a self-training method based on a CTC alignment algorithm that matches full transcriptions to text line images using dynamic programming and model output probabilities trained with the CTC loss. Our approach improves performance (e.g., by 1.1 percentage points CER with PyLaia) and increases alignment accuracy. Interestingly, we find that weaker models yield more accurate alignments, enabling an iterative training strategy. We release a new manually corrected subset of 100 pages from the Bullinger dataset, along with our code and benchmarks. Our approach can be applied iteratively to further improve the CER as well as the alignment quality for text recognition pipelines. Code and data are available via https://github.com/andreas-fischer-unifr/nntp.

cs.CV↗

arXiv2025

HiSTM: Hierarchical Spatiotemporal Mamba for Cellular Traffic Forecasting

Cellular traffic forecasting is essential for network planning, resource allocation, or load-balancing traffic across cells. However, accurate forecasting is difficult due to intricate spatial and temporal patterns that exist due to the mobility of users. Existing AI-based traffic forecasting models often trade-off accuracy and computational efficiency. We present Hierarchical SpatioTemporal Mamba (HiSTM), which combines a dual spatial encoder with a Mamba-based temporal module and attention mechanism. HiSTM employs selective state space methods to capture spatial and temporal patterns in network traffic. In our evaluation, we use a real-world dataset to compare HiSTM against several baselines, showing a 29.4% MAE improvement over the STN baseline while using 94% fewer parameters. We show that the HiSTM generalizes well across different datasets and improves in accuracy over longer time-horizons.

cs.NI↗

arXiv2025

Enhancing Spatiotemporal Networks with xLSTM: A Scalar LSTM Approach for Cellular Traffic Forecasting

Accurate spatiotemporal traffic forecasting is vital for intelligent resource management in 5G and beyond. However, conventional AI approaches often fail to capture the intricate spatial and temporal patterns that exist, due to e.g., the mobility of users. We introduce a lightweight, dual-path Spatiotemporal Network that leverages a Scalar LSTM (sLSTM) for efficient temporal modeling and a three-layer Conv3D module for spatial feature extraction. A fusion layer integrates both streams into a cohesive representation, enabling robust forecasting. Our design improves gradient stability and convergence speed while reducing prediction error. Evaluations on real-world datasets show superior forecast performance over ConvLSTM baselines and strong generalization to unseen regions, making it well-suited for large-scale, next-generation network deployments. Experimental evaluation shows a 23% MAE reduction over ConvLSTM, with a 30% improvement in model generalization.

cs.LG↗

arXiv2025

Synthetic Data Augmentation for Table Detection: Re-evaluating TableNet's Performance with Automatically Generated Document Images

Document pages captured by smartphones or scanners often contain tables, yet manual extraction is slow and error-prone. We introduce an automated LaTeX-based pipeline that synthesizes realistic two-column pages with visually diverse table layouts and aligned ground-truth masks. The generated corpus augments the real-world Marmot benchmark and enables a systematic resolution study of TableNet. Training TableNet on our synthetic data achieves a pixel-wise XOR error of 4.04% on our synthetic test set with a 256x256 input resolution, and 4.33% with 1024x1024. The best performance on the Marmot benchmark is 9.18% (at 256x256), while cutting manual annotation effort through automation.

cs.CV↗

arXiv2024

Zero-Shot Prompting and Few-Shot Fine-Tuning: Revisiting Document Image Classification Using Large Language Models

Classifying scanned documents is a challenging problem that involves image, layout, and text analysis for document understanding. Nevertheless, for certain benchmark datasets, notably RVL-CDIP, the state of the art is closing in to near-perfect performance when considering hundreds of thousands of training samples. With the advent of large language models (LLMs), which are excellent few-shot learners, the question arises to what extent the document classification problem can be addressed with only a few training samples, or even none at all. In this paper, we investigate this question in the context of zero-shot prompting and few-shot model fine-tuning, with the aim of reducing the need for human-annotated training samples as much as possible.

cs.CV↗