Lydia Müller

Julian Oestreich, Lydia Müller

We present a comprehensive evaluation of structured decoding for text-to-table generation with large language models (LLMs). While previous work has primarily focused on unconstrained generation of tables, the impact of enforcing structural constraints during generation remains underexplored. We systematically compare schema-guided (structured) decoding to standard one-shot prompting across three diverse benchmarks - E2E, Rotowire, and Livesum - using open-source LLMs of up to 32B parameters, assessing the performance of table generation approaches in resource-constrained settings. Our experiments cover a wide range of evaluation metrics at cell, row, and table levels. Results demonstrate that structured decoding significantly enhances the validity and alignment of generated tables, particularly in scenarios demanding precise numerical alignment (Rotowire), but may degrade performance in contexts involving densely packed textual information (E2E) or extensive aggregation over lengthy texts (Livesum). We further analyze the suitability of different evaluation metrics and discuss the influence of model size.

Christopher Schröder, Lydia Müller, Andreas Niekler et al.

We introduce small-text, an easy-to-use active learning library, which offers pool-based active learning for single- and multi-label text classification in Python. It features numerous pre-implemented state-of-the-art query strategies, including some that leverage the GPU. Standardized interfaces allow the combination of a variety of classifiers, query strategies, and stopping criteria, facilitating a quick mix and match, and enabling a rapid and convenient development of both active learning experiments and applications. With the objective of making various classifiers and query strategies accessible for active learning, small-text integrates several well-known machine learning libraries, namely scikit-learn, PyTorch, and Hugging Face transformers. The latter integrations are optionally installable extensions, so GPUs can be used but are not required. Using this new library, we investigate the performance of the recently published SetFit training paradigm, which we compare to vanilla transformer fine-tuning, finding that it matches the latter in classification accuracy while outperforming it in area under the curve. The library is available under the MIT License at https://github.com/webis-de/small-text, in version 1.3.0 at the time of writing.

Christopher Schröder, Kim Bürgl, Yves Annanias et al.

Open pit mines left many regions worldwide inhospitable or uninhabitable. To put these regions back into use, entire stretches of land must be renaturalized. For the sustainable subsequent use or transfer to a new primary use, many contaminated sites and soil information have to be permanently managed. In most cases, this information is available in the form of expert reports in unstructured data collections or file folders, which in the best case are digitized. Due to size and complexity of the data, it is difficult for a single person to have an overview of this data in order to be able to make reliable statements. This is one of the most important obstacles to the rapid transfer of these areas to after-use. An information-based approach to this issue supports fulfilling several Sustainable Development Goals regarding environment issues, health and climate action. We use a stack of Optical Character Recognition, Text Classification, Active Learning and Geographic Information System Visualization to effectively mine and visualize this information. Subsequently, we link the extracted information to geographic coordinates and visualize them using a Geographic Information System. Active Learning plays a vital role because our dataset provides no training data. In total, we process nine categories and actively learn their representation in our dataset. We evaluate the OCR, Active Learning and Text Classification separately to report the performance of the system. Active Learning and text classification results are twofold: Whereas our categories about restrictions work sufficient ($>$.85 F1), the seven topic-oriented categories were complicated for human coders and hence the results achieved mediocre evaluation scores ($<$.70 F1).