In Search of Needles in a 10M Haystack: Recurrent Memory Finds What LLMs Miss

Yury Kuratov, Aydar Bulatov, Petr Anokhin, Dmitry Sorokin, Artyom Sorokin, MIKHAIL BURTSEV

In Search of Needles in a 10M Haystack: Recurrent Memory Finds What LLMs Miss: 42 upvotes on Hugging Face Daily Papers, #2 of 12 papers on 2024-02-19. Day-by-day upvote history.

This paper addresses the challenge of processing long documents using generative transformer models. To evaluate different approaches, we introduce BABILong, a new benchmark designed to assess model capabilities in extracting and processing distributed facts within extensive texts. Our evaluation, which includes benchmarks for GPT-4 and RAG, reveals that common methods are effective only for sequences up to 10^4 elements. In contrast, fine-tuning GPT-2 with recurrent memory augmentations enables it to handle tasks involving up to 10^7 elements. This achievement marks a substantial leap, as it is by far the longest input processed by any open neural network model to date, demonstrating a significant improvement in the processing capabilities for long sequences.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.