Papers by Ali Safaya
Neurocache: Efficient Vector Retrieval for Long-range Language Modeling (2024.naacl-long)
Copied to clipboard
| Challenge: | Recent research shows that retrieval-augmented models with shorter contexts (4K tokens) can match the performance of models with longer contexts (16K/32K token) |
| Approach: | They introduce an approach to extend the effective context size of large language models by using an external vector cache to store past states. |
| Outcome: | The proposed method improves on models trained from scratch and pre-trained models. |
Mukayese: Turkish NLP Strikes Back (2022.findings-acl)
Copied to clipboard
| Challenge: | Having sufficient resources for language X lifts it from the under-resourced languages class, but not necessarily from the researched class. |
| Approach: | They propose a set of NLP benchmarks for the Turkish language that contains several NLP tasks. |
| Outcome: | The proposed benchmarks outperform previous work significantly in the Turkish language. |