
Improving NLI Robustness via Targeted Fine-Tuning
Breaking and fixing the SNLI benchmark.
View on GitHub
Fine-Tuning TinyLlama for Medical QA
LoRA + CoT rejection sampling to answer layperson medical questions.
View on GitHub
Breaking and fixing the SNLI benchmark.
View on GitHub
LoRA + CoT rejection sampling to answer layperson medical questions.
View on GitHub