Memory constraints remain a critical bottleneck in the deployment of large-scale AI models. Parameter sharing across network depth reduces model storage, but repeatedly applying an identical transformation limits flexibility across layers. Inspired by time--memory trade-offs in classical algorithms, we introduce the Li...
This work develops a geometric framework for characterizing the distributional structure of empirical datasets by quantifying their deviation from the Gaussian family under the geometry induced by optimal transport theory and develops a numerical approximation method for the proposed quantities based on empirical optim...
The Linear Reusable Neural Bases Architecture (LRNBA) is introduced, a novel framework aimed at improving parameter efficiency and reducing memory cost and inspired by recurrent neural network designs.
It is shown that paragraph supervision enables effective use of long token sequences, whereas caption-only training degrades beyond 60 tokens, and paragraph supervision consistently benefits long-description benchmarks and hard negatives prove detrimental in text-only fine-tuning.
Mahyar Ghazanfari, Amin Tabrizian, Arsyi Aziz et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
