TY  - RPRT
TI  - Beyond Monolithic Rewards: A Hybrid and Multi-Aspect Reward Optimization for MLLM Alignment
AU  - Radha Gulhane
AU  - Sathish Reddy Indurthi
PY  - 2025
UR  - https://arxiv.org/abs/2510.05283
ID  - 2510.05283
ER  -