Debiasing-DPO reduces bias to spurious social contexts by 84% and improves predictive accuracy by 52% on average for LLMs evaluating U.S. classroom transcripts.
ISBN 9798400706332
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
PEFT fine-tuning of Code Llama yields feedback on student Java bugs that students judge equal to ChatGPT and better than prompt engineering, using BLEU/ROUGE/BERTScore plus human ratings.
citing papers explorer
-
Mitigating LLM biases toward spurious social contexts using direct preference optimization
Debiasing-DPO reduces bias to spurious social contexts by 84% and improves predictive accuracy by 52% on average for LLMs evaluating U.S. classroom transcripts.
-
Fine-Tuning Models for Automated Code Review Feedback
PEFT fine-tuning of Code Llama yields feedback on student Java bugs that students judge equal to ChatGPT and better than prompt engineering, using BLEU/ROUGE/BERTScore plus human ratings.