Parameter-efficient fine-tuning (LoRA family)
MTLoRA: A Low-Rank Adaptation Approach for Efficient Multi-Task Learning
Cited as a baseline — critiqued by newer work, not yet beaten on a benchmark here
1 papers critique it · 0 beat it on benchmarks
Verbatim critique sentences, each from a paper that cites MTLoRA as a baseline.
However, our experimental results indicate that these models are less effective in multi-modal fusion.
Recent methods in the same sub-problem, not yet superseded in the knowledge base — arXiv benchmark leaders, not vetted production recommendations.