MUCnoHARM@GermEval Shared Task 2026: Retrieval-based In-Context Learning for Defamatory Offences, and Where It Falls Short
Not on the map. Why?
Abstract (arXiv)
With hate speech being ubiquitous online, automatic detection is crucial, in particular when it comes to criminally relevant social media posts. We study a variety of retrieval-based in-context learning (RetICL) strategies for detecting defamatory offences under SS 185-187 StGB (the subject of GermEval 2026 Subtask 4). Few-shot prompting beats zero-shot, but retrieval-based approaches offer only marginal gains over random demonstrations, and even fall behind an optimised static set of demonstrations. Providing concrete legal knowledge helps, yet model choice outweighs every other system choice. Models over-predict criminal relevance while still missing 26-57% of criminally relevant posts, suiting them for triage rather than autonomous moderation.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 12 of its 44 references that are in the map. First posted on arXiv 9 Sep 2026.
Builds on
12 of 44 references.
References · 12
- 2021Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivitycited by 1,135
- 2023Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formattingcited by 817
- 2023Legal Syllogism Prompting: Teaching Large Language Models for Legal Judgment Predictioncited by 38
- 2023Sensitivity and Robustness of Large Language Models to Prompt Template in Japanese Text Classification Taskscited by 25
- 2024Enabling Discriminative Reasoning in LLMs for Legal Judgment Predictioncited by 17
- 2024Improving Adversarial Data Collection by Supporting Annotators: Lessons from GAHD, a German Hate Speech Datasetcited by 13
- 2020A Legal Approach to Hate Speech: Operationalizing the EU's Legal Framework against the Expression of Hatred as an NLP Taskcited by 11
- 2024Making FETCH! Happen: Finding Emergent Dog Whistles Through Common Habitatscited by 6
- 2025Conditioning Large Language Models on Legal Systems? Detecting Punishable Hate Speechcited by 2
- 2025The Unintended Trade-off of AI Alignment:Balancing Hallucination Mitigation and Safety in LLMscited by 2
- 2025Confident, Calibrated, or Complicit: Safety Alignment and Ideological Bias in LLM Hate Speech Detectioncited by 1
- 2026Beyond Imperfect Alternatives with Rulemapping: A Neuro-Symbolic Case Study on Online Hate Speechcited by 1
Built on by
Not cited yet.