Forgetting Curve: A Reliable Method for Evaluating Memorization Capability for Long-Context Models

Xinyu Liu, Runsong Zhao, Pengcheng Huang, Chunyang Xiao, Bei Li, Jingang Wang, Tong Xiao, JingBo Zhu


Abstract
Numerous recent works target to extend effective context length for language models and various methods, tasks and benchmarks exist to measure model’s effective memory length. However, through thorough investigations, we find limitations for currently existing evaluations on model’s memory. We provide an extensive survey for limitations in this work and propose a new method called forgetting curve to measure the memorization capability of long-context models. We show that forgetting curve has the advantage of being robust to the tested corpus and the experimental settings, of not relying on prompt and can be applied to any model size. We apply our forgetting curve to a large variety of models involving both transformer and RNN/SSM based architectures. Our measurement provides empirical evidence for the effectiveness of transformer extension techniques while raises questions for the effective length of RNN/SSM based models. We also examine the difference between our measurement and existing benchmarks as well as popular metrics for various models.
Anthology ID:
2024.emnlp-main.269
Volume:
Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2024
Address:
Miami, Florida, USA
Editors:
Yaser Al-Onaizan, Mohit Bansal, Yun-Nung Chen
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
4667–4682
Language:
URL:
https://aclanthology.org/2024.emnlp-main.269/
DOI:
10.18653/v1/2024.emnlp-main.269
Bibkey:
Cite (ACL):
Xinyu Liu, Runsong Zhao, Pengcheng Huang, Chunyang Xiao, Bei Li, Jingang Wang, Tong Xiao, and JingBo Zhu. 2024. Forgetting Curve: A Reliable Method for Evaluating Memorization Capability for Long-Context Models. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 4667–4682, Miami, Florida, USA. Association for Computational Linguistics.
Cite (Informal):
Forgetting Curve: A Reliable Method for Evaluating Memorization Capability for Long-Context Models (Liu et al., EMNLP 2024)
Copy Citation:
PDF:
https://aclanthology.org/2024.emnlp-main.269.pdf

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy