MARS Accepted to NeurIPS'26
MARS addresses imbalanced training dynamics in MLLM finetuning by finding the optimal rank pair, leveraging our proposed dual scaling laws to guide the search. See you at Sydney 🦘🐨🇦🇺~!
MARS addresses imbalanced training dynamics in MLLM finetuning by finding the optimal rank pair, leveraging our proposed dual scaling laws to guide the search. See you at Sydney 🦘🐨🇦🇺~!