Using transformers to improve answer retrieval for legal questions
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Transformer architectures such as BERT, XLNet, and others are frequently used in the field of natural language processing. Transformers have achieved state-of-the-art performance in tasks such as text classification, passage summarization, machine translation, and question answering. Efficient hosting of transformer models, however, is a difficult task because of their large size and high latency. In this work, we describe how we deploy a RoBERTa Base question answer classification model in a production environment. We also compare the answer retrieval performance of a RoBERTa Base classifier against a traditional machine learning model in the legal domain by measuring the performance difference between a trained linear SVM on the publicly available PRIVACYQA dataset. We show that RoBERTa achieves a 31% improvement in F1-score and a 41% improvement in Mean Reciprocal Rank over the traditional SVM.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
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 it