Measuring & Mitigating Over-Alignment for LLMs in Multilingual Criminal Law Courts
While the wider applicability of LLMs in the legal field is currently debated due to their reliability and the gravity of any errors, narrow uses with well-understood and mitigated risks have emerged. Notably the Swiss Federal Supreme Court uses small on-premises models for tentative translations and short-passage summarization across the four official languages. However, such usage is challenging in the context of Criminal Law. Since rulings and cases employees work on routinely can contain detailed descriptions of violent and sexual offenses, their legitimate work is compromised by refusals and disclaimers due to the activation of model guardrails (over-alignment). To measure this phenomenon, we introduce TF-RefusalBench, a multilingual benchmark for criminal-law translation and summarization derived from public Swiss Supreme Court rulings. TF-RefusalBench contains 5,200 total prompts across French, German, Italian, and English, corresponding to common task prompts and passages likely to trigger refusal. We then use TF-RefusalBench to show that over-alignment is a multifaceted phenomenon, influenced by the model and the prompt and text languages being processed, and that its impact cannot be evaluated solely from an over-refusal perspective, given the disclaimer's impact on task faithfulness. Finally, we evaluate approaches to enable on-premises LLMs for Criminal Law Tasks, demonstrating that while prompting can be effective, abliteration (refusal directions ablation) eliminates refusal with minimal impact on task performance.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Mending the Holes: Mitigating Reward Hacking in Reinforcement Learning for Multilingual Translation
Large Language Models (LLMs) have demonstrated remarkable capability in machine translation on high-resource language pairs, yet their performance on low-resource translation still lags behind. Existing post-training met…
Reinforcement LearningMachine TranslationWord AlignmentA Data-Efficient Path to Multilingual LLMs: Language Expansion via Post-training PARAM$Δ$ Integration into Upcycled MoE
Expanding Large Language Models~(LLMs) to new languages is a costly endeavor, demanding extensive Continued Pre-Training~(CPT) and data-intensive alignment. While recent data-free merging techniques attempt to bypass ali…
Language AcquisitionAlignX: Advancing Multilingual Large Language Models with Multilingual Representation Alignment
Multilingual large language models (LLMs) possess impressive multilingual understanding and generation capabilities. However, their performance and cross-lingual alignment often lag for non-dominant languages. A common s…
Aligning to What? Limits to RLHF Based Alignment
Reinforcement Learning from Human Feedback (RLHF) is increasingly used to align large language models (LLMs) with human preferences. However, the effectiveness of RLHF in addressing underlying biases remains unclear. Thi…
Getting More from Less: Large Language Models are Good Spontaneous Multilingual Learners
Recently, Large Language Models (LLMs) have shown impressive language capabilities. While most of the existing LLMs have very unbalanced performance across different languages, multilingual alignment based on translation…
Translation