We present a comprehensive study of LLMs for argument mining, comparing four learning paradigms against a strong encoder-only baseline across three heterogeneous corpora to assess cross-domain generalisation. We evaluate in-context learning, supervised fine-tuning (SFT), and cold-start reinforcement fine-tuning (RFT) via GRPO. Our results demonstrate that BERT-style encoder-only transformers consistently outperform much larger decoder-only LLMs across all settings. Further mechanistic analysis reveals that the central bottleneck is not knowledge but expression: linear probing of frozen hidden states recovers argument relations at 69–80% accuracy from base models, yet edge generation remains poor across all paradigms. Attention analysis confirms a universal structural blindness across 27 tested conditions.

This paper presents the UZH Shared Task at the 13th Workshop on Argument Mining and Reasoning, co-located with ACL 2026, which focuses on reconstructing argumentative structure in highly formal legal-political texts, namely United Nations resolutions and recommendations. The shared task addresses the challenge of recovering paragraph-level reasoning patterns from the fairly formulaic structure of international decision-making records. It comprises two subtasks: (1) paragraph classification, where systems identify paragraph type (preambular or operative) and assign one or more thematic tags, and (2) argumentative relation prediction, where systems infer links between paragraphs and label them with relation types such as supporting, contradictive, complemental, and modifying. The data is provided in French, together with English translations. It includes a test set of resolutions and recommendations from the UNESCO International Conference on Education, annotated at paragraph level, as well as unlabeled training resolutions from the United Nations. The task restricts submissions to open-weight language models with at most 8 billion parameters and requires demonstration of models’ reasoning capabilities. By launching this task, we aim to advance research on argument mining in formal institutional discourse, particularly in multilingual, policy-oriented documents.