Skip to main content
Search

# LoRA-GT

1 related articles

Subscribe to updates
  1. arXiv

    Safety judges can now carry generalisation guarantees: LoRA-GT gives the tightest certificate in all 40 model–dataset–loss cases

    For large language model judges that classify prompt–response pairs from a fixed generator, a PAC-Bayes certification pipeline yields non-vacuous and sometimes tight risk bounds without holding out a separate validation set, covering both 0–1 classification risk and Brier risk. It quantises and losslessly compresses the adaptor produced by standard parameter-efficient fine-tuning and uses its description length as the complexity term; the new LoRA-GT shares one quantisation scale across layers and gives the tightest certificate in all 40 model–dataset–loss cases across five guardrail models and four benchmarks. Certificate tightness tracks adaptation complexity more than base model scale.

You’re all caught up