TY  - RPRT
TI  - Oracle-RLAIF: An Improved Multi-Modal Fine-Tuning Framework through Reinforcement Learning from Ranking Feedback
AU  - Shi, Derek [University of California Los Angeles]
AU  - Glatt, Ruben [Lawrence Livermore National Laboratory]
AU  - Klymko, Christine [Lawrence Livermore National Laboratory]
AU  - Mohole, Shubham
AU  - Kushwaha, Shashank
AU  - Choi, Hongjun [Lawrence Livermore National Laboratory]
AU  - Sakla, Wesam [Lawrence Livermore National Laboratory]
AU  - Leno Da Silva, Leno [Lawrence Livermore National Laboratory]
PY  - 2026
DO  - 10.2172/3389288
UR  - https://www.osti.gov/biblio/3389288
ID  - 3389288
ER  -