Aligning Language Models for Icelandic Legal Text Summarization
This work addresses the challenge of aligning language models with domain-specific standards for legal professionals in Iceland, though it is incremental as it builds on existing preference training methods.
The study tackled the problem of generating Icelandic legal text summaries by applying preference-based training techniques like Reinforcement Learning from Human Feedback and Direct Preference Optimization, finding that these methods improved legal accuracy compared to standard fine-tuning but did not significantly enhance overall Icelandic language quality.
The integration of language models in the legal domain holds considerable promise for streamlining processes and improving efficiency in managing extensive workloads. However, the specialized terminology, nuanced language, and formal style of legal texts can present substantial challenges. This study examines whether preference-based training techniques, specifically Reinforcement Learning from Human Feedback and Direct Preference Optimization, can enhance models' performance in generating Icelandic legal summaries that align with domain-specific language standards and user preferences. We compare models fine-tuned with preference training to those using conventional supervised learning. Results indicate that preference training improves the legal accuracy of generated summaries over standard fine-tuning but does not significantly enhance the overall quality of Icelandic language usage. Discrepancies between automated metrics and human evaluations further underscore the importance of qualitative assessment in developing language models for the legal domain.