English
Related papers

Related papers: Continual Memorization of Factoids in Language Mod…

200 papers

Machine unlearning aims to remove the influence of specific training data from a model without requiring full retraining. This capability is crucial for ensuring privacy, safety, and regulatory compliance. Therefore, verifying whether a…

Computation and Language · Computer Science 2025-11-07 Liran Cohen , Yaniv Nemcovesky , Avi Mendelson

Large language models (LLMs) have shown great capabilities in various tasks but also exhibited memorization of training data, raising tremendous privacy and copyright concerns. While prior works have studied memorization during…

Artificial Intelligence · Computer Science 2024-02-26 Shenglai Zeng , Yaxin Li , Jie Ren , Yiding Liu , Han Xu , Pengfei He , Yue Xing , Shuaiqiang Wang , Jiliang Tang , Dawei Yin

Large language models may encounter factual knowledge during pre-training yet fail to reliably use that knowledge after fine-tuning. Despite growing empirical evidence that MLP layers store factual associations and fine-tuning affects…

Machine Learning · Computer Science 2026-05-19 Ruichen Xu , Kexin Chen

Modern language models are powerful, but typically static after deployment. A major obstacle to building models that continually learn over time is catastrophic forgetting, where updating on new data erases previously acquired capabilities.…

Computation and Language · Computer Science 2025-10-20 Jessy Lin , Luke Zettlemoyer , Gargi Ghosh , Wen-Tau Yih , Aram Markosyan , Vincent-Pierre Berges , Barlas Oğuz

Catastrophic forgetting (CF) occurs when a neural network loses the information previously learned while training on a set of samples from a different distribution, i.e., a new task. Existing approaches have achieved remarkable results in…

Machine Learning · Computer Science 2022-09-13 Jary Pomponi , Simone Scardapane , Aurelio Uncini

Large language models (LLMs) and multimodal models (MMs) have exhibited impressive capabilities in various domains, particularly in general language understanding and visual reasoning. However, these models, trained on massive data, may not…

Computation and Language · Computer Science 2024-12-19 Xinbo Wu , Max Hartman , Vidhata Arjun Jayaraman , Lav R. Varshney

Language models (LMs) can "memorize" information, i.e., encode training data in their weights in such a way that inference-time queries can lead to verbatim regurgitation of that data. This ability to extract training data can be…

This paper considers continual learning of large-scale pretrained neural machine translation model without accessing the previous training data or introducing model separation. We argue that the widely used regularization-based methods,…

Computation and Language · Computer Science 2022-11-07 Shuhao Gu , Bojie Hu , Yang Feng

Memorization, or the tendency of large language models (LLMs) to output entire sequences from their training data verbatim, is a key concern for safely deploying language models. In particular, it is vital to minimize a model's memorization…

Computation and Language · Computer Science 2023-06-02 Stella Biderman , USVSN Sai Prashanth , Lintang Sutawika , Hailey Schoelkopf , Quentin Anthony , Shivanshu Purohit , Edward Raff

Catastrophic Forgetting (CF) means models forgetting previously acquired knowledge when learning new data. It compromises the effectiveness of large language models (LLMs) during fine-tuning, yet the underlying causes have not been…

Computation and Language · Computer Science 2024-06-10 Hongyu Li , Liang Ding , Meng Fang , Dacheng Tao

Can we localize the weights and mechanisms used by a language model to memorize and recite entire paragraphs of its training data? In this paper, we show that while memorization is spread across multiple layers and model components,…

Computation and Language · Computer Science 2024-04-01 Niklas Stoehr , Mitchell Gordon , Chiyuan Zhang , Owen Lewis

Large language models (LLMs) generate fluent text across a wide range of tasks, but the fabrication of non-existent academic citations remains a critical and well-documented failure mode. Building on prior work that frames hallucination and…

Computation and Language · Computer Science 2026-05-06 Junichiro Niimi

Large language models accumulate vast knowledge during pre-training, yet the dynamics governing this acquisition remain poorly understood. This work investigates the learning dynamics of language models on a synthetic factual recall task,…

Computation and Language · Computer Science 2025-07-25 Nicolas Zucchet , Jörg Bornschein , Stephanie Chan , Andrew Lampinen , Razvan Pascanu , Soham De

Deep neural networks suffer from catastrophic forgetting, where performance on previous tasks degrades after training on a new task. This issue arises due to the model's tendency to overwrite previously acquired knowledge with new…

Machine Learning · Computer Science 2025-12-02 Lama Alssum , Hasan Abed Al Kader Hammoud , Motasem Alfarra , Juan C Leon Alcazar , Bernard Ghanem

Continual Pre-Training (CPT) is essential for enabling Language Models (LMs) to integrate new knowledge without erasing old. While classical CPT techniques like data replay have become the standard paradigm, the mechanisms underlying how…

Computation and Language · Computer Science 2026-05-12 Haoyu Wang , Yifan Shang , Zhongxiang Sun , Weijie Yu , Xiao Zhang , Jun Xu

Large language models (LLMs) achieve good performance on challenging reasoning benchmarks, yet could also make basic reasoning mistakes. This contrasting behavior is puzzling when it comes to understanding the mechanisms behind LLMs'…

Computation and Language · Computer Science 2025-03-05 Chulin Xie , Yangsibo Huang , Chiyuan Zhang , Da Yu , Xinyun Chen , Bill Yuchen Lin , Bo Li , Badih Ghazi , Ravi Kumar

Foundation Models (FMs) have become the hallmark of modern AI, however, these models are trained on massive data, leading to financially expensive training. Updating FMs as new data becomes available is important, however, can lead to…

Machine Learning · Computer Science 2024-04-22 James Seale Smith , Lazar Valkov , Shaunak Halbe , Vyshnavi Gutta , Rogerio Feris , Zsolt Kira , Leonid Karlinsky

Forgetting is often seen as an unwanted characteristic in both human and machine learning. However, we propose that forgetting can in fact be favorable to learning. We introduce "forget-and-relearn" as a powerful paradigm for shaping the…

Machine Learning · Computer Science 2022-02-02 Hattie Zhou , Ankit Vani , Hugo Larochelle , Aaron Courville

Large language models (LLMs) are often fine-tuned for use on downstream tasks, though this can degrade capabilities learned during previous training. This phenomenon, often referred to as catastrophic forgetting, has important potential…

Computation and Language · Computer Science 2024-12-30 Megan Ung , Alicia Sun , Samuel J. Bell , Bhaktipriya Radharapu , Levent Sagun , Adina Williams

As the number of large language models (LLMs) released to the public grows, there is a pressing need to understand the safety implications associated with these models learning from third-party custom finetuning data. We explore the…

Computation and Language · Computer Science 2024-07-04 Jiachen Zhao , Zhun Deng , David Madras , James Zou , Mengye Ren