JudgeMeNot: Personalizing Large Language Models to Emulate Judicial Reasoning in Hebrew
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918456590336000 |
|---|---|
| author | Razumenko, Itay Sturm, Arnon Grinberg, Nir |
| author_facet | Razumenko, Itay Sturm, Arnon Grinberg, Nir |
| contents | Despite significant advances in large language models, personalizing them for individual decision-makers remains an open problem. Here, we introduce a synthetic-organic supervision pipeline that transforms raw judicial decisions into instruction-tuning data, enabling parameter-efficient fine-tuning of personalized models for individual judges in low-resource settings. We compare our approach to state-of-the-art personalization techniques across three different tasks and settings. The results show that Causal Language Modeling followed by synthetically generated instruction-tuning significantly outperforms all other baselines, providing significant improvements across lexical, stylistic, and semantic similarity. Notably, our model-generated outputs are indistinguishable from the reasoning of human judges, highlighting the viability of efficient personalization, even in low-resource settings. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_18041 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | JudgeMeNot: Personalizing Large Language Models to Emulate Judicial Reasoning in Hebrew Razumenko, Itay Sturm, Arnon Grinberg, Nir Computation and Language Computers and Society Despite significant advances in large language models, personalizing them for individual decision-makers remains an open problem. Here, we introduce a synthetic-organic supervision pipeline that transforms raw judicial decisions into instruction-tuning data, enabling parameter-efficient fine-tuning of personalized models for individual judges in low-resource settings. We compare our approach to state-of-the-art personalization techniques across three different tasks and settings. The results show that Causal Language Modeling followed by synthetically generated instruction-tuning significantly outperforms all other baselines, providing significant improvements across lexical, stylistic, and semantic similarity. Notably, our model-generated outputs are indistinguishable from the reasoning of human judges, highlighting the viability of efficient personalization, even in low-resource settings. |
| title | JudgeMeNot: Personalizing Large Language Models to Emulate Judicial Reasoning in Hebrew |
| topic | Computation and Language Computers and Society |
| url | https://arxiv.org/abs/2604.18041 |