Veronika Solopova

Alignment has become the dominant paradigm for governing large language models (LLMs), implemented through post-training optimization, safety fine-tuning, and inference-time steering. These approaches are commonly assumed to translate normative intent into reliable system behavior. As LLMs are increasingly deployed as large-scale decision-support and information systems, however, the limits of this assumption have become evident. This chapter analyzes alignment from a systems and governance perspective, examining how alignment and steering mechanisms interact with scale, infrastructure, and deployment dynamics. We argue that contemporary approaches primarily regulate observable output behavior through probabilistic narrative control rather than enforcing durable epistemic constraints. While effective at shaping refusals, explanations, and interaction norms, they do not guarantee robustness, truthfulness, or stability under distribution shift or strategic pressure. As a result, aligned systems may appear compliant while remaining epistemically fragile. Current legal and ethical governance frameworks, which emphasize documentation, transparency, and design-time intent, struggle to address this gap because they operate upstream of the deployment-time mechanisms that shape behavior in practice. Situating alignment within the global political economy of AI, the chapter shows how capital concentration, infrastructural lock-in, and competitive deployment amplify these limitations by redistributing effective control toward actors with access to compute, data, and platforms. These pressures feed back into model behavior, producing internal failure modes such as sycophancy, over-alignment, and calibration collapse. We conclude by reframing alignment as a bounded interface-level control mechanism, highlighting the need for system-level oversight beyond training-time alignment alone.

This notebook reports the XplaiNLP submission to the CheckThat! 2025 shared task on multilingual subjectivity detection. We evaluate two approaches: (1) supervised fine-tuning of transformer encoders, EuroBERT, XLM-RoBERTa, and German-BERT, on monolingual and machine-translated training data; and (2) zero-shot prompting using two LLMs: o3-mini for Annotation (rule-based labelling) and gpt-4.1-mini for DoubleDown (contrastive rewriting) and Perspective (comparative reasoning). The Annotation Approach achieves 1st place in the Italian monolingual subtask with an F_1 score of 0.8104, outperforming the baseline of 0.6941. In the Romanian zero-shot setting, the fine-tuned XLM-RoBERTa model obtains an F_1 score of 0.7917, ranking 3rd and exceeding the baseline of 0.6461. The same model also performs reliably in the multilingual task and improves over the baseline in Greek. For German, a German-BERT model fine-tuned on translated training data from typologically related languages yields competitive performance over the baseline. In contrast, performance in the Ukrainian and Polish zero-shot settings falls slightly below the respective baselines, reflecting the challenge of generalization in low-resource cross-lingual scenarios.

Incels (involuntary celibates) are part of a broader misogynistic culture known as the manosphere. Some communities within the manosphere, including incels, promote gender-based violence through misogynistic rhetoric and ideology. Incels are men who struggle to form romantic relationships and thus seek solace in online forums to find a sense of purpose and community. The community is organised around an ideology and a hierarchical classification of members. This paper presents a computational linguistic analysis of the utterances made within the community. We analyse the linguistic patterns of six different ranks of users: paragons, high-rank users, low-rank users, greycels, banned users, and self-banned/deleted users. We conducted an analysis of the sentiments expressed between ranks and of the affective attitude of posters when discussing a variety of topics. We then we analysed linguistic features, also conditioned on the rank of the poster. These analyses reveal qualitative differences between users of different ranks in how they express themselves and their sentiments. Finally, we trained models based on the linguistic features of users’ posts and show that already based on this information it is possible to predict the rank of users. Overall, our results indicate that the incel community has developed distinctive linguistic patterns to disseminate their ideology and that these linguistic patterns vary according to the rank of an incel within the community. This methodology offers a robust framework for studying similar echo chamber-like communities.

We introduce VIBA, a novel approach for explainable video classification by adapting Information Bottlenecks for Attribution (IBA) to video sequences. While most traditional explainability methods are designed for image models, our IBA framework addresses the need for explainability in temporal models used for video analysis. To demonstrate its effectiveness, we apply VIBA to video deepfake detection, testing it on two architectures: the Xception model for spatial features and a VGG11-based model for capturing motion dynamics through optical flow. Using a custom dataset that reflects recent deepfake generation techniques, we adapt IBA to create relevance and optical flow maps, visually highlighting manipulated regions and motion inconsistencies. Our results show that VIBA generates temporally and spatially consistent explanations, which align closely with human annotations, thus providing interpretability for video classification and particularly for deepfake detection.

Yurii Shynkarov, Veronika Solopova, Vera Schmitt. Proceedings of the Fourth Ukrainian Natural Language Processing Workshop (UNLP 2025). 2025.

Many European citizens become targets of the Kremlin propaganda campaigns, aiming to minimise public support for Ukraine, foster a climate of mistrust and disunity, and shape elections (Meister, 2022). To address this challenge, we developed ''Check News in 1 Click'', the first NLP-empowered pro-Kremlin propaganda detection application available in 7 languages, which provides the lay user with feedback on their news, and explains manipulative linguistic features and keywords. We conducted a user study, analysed user entries and models' behaviour paired with questionnaire answers, and investigated the advantages and disadvantages of the proposed interpretative solution.

Veronika Solopova, Viktoriia Herman, Christoph Benzmüller, Tim Landgraf. Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics: System Demonstrations. 2024.