A Comparative Study of Prompting Strategies for Legal Text Classification
Bibliographic record
Abstract
In this study, we explore the performance of large language models (LLMs) using different prompt engineering approaches in the context of legal text classification.Prior research has demonstrated that various prompting techniques can improve the performance of a diverse array of tasks done by LLMs.However, in this research, we observe that professional documents, and in particular legal documents, pose unique challenges for LLMs.We experiment with several LLMs and various prompting techniques, including zero/few-shot prompting, prompt ensembling, chain-of-thought, and activation fine-tuning and compare the performance on legal datasets.Although the new generation of LLMs and prompt optimization techniques have been shown to improve generation and understanding of generic tasks, our findings suggest that such improvements may not readily transfer to other domains.Specifically, experiments indicate that not all prompting approaches and models are well-suited for the legal domain which involves complexities such as long documents and domain-specific language.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.008 | 0.033 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.002 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.002 | 0.005 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.002 | 0.003 |
| Insufficient payload (model declined to judge) | 0.003 | 0.003 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".