Recite, Reconstruct, Recollect: Memorization in LMs as a Multifaceted Phenomenon
- URL: http://arxiv.org/abs/2406.17746v1
- Date: Tue, 25 Jun 2024 17:32:16 GMT
- Title: Recite, Reconstruct, Recollect: Memorization in LMs as a Multifaceted Phenomenon
- Authors: USVSN Sai Prashanth, Alvin Deng, Kyle O'Brien, Jyothir S V, Mohammad Aflah Khan, Jaydeep Borkar, Christopher A. Choquette-Choo, Jacob Ray Fuehne, Stella Biderman, Tracy Ke, Katherine Lee, Naomi Saphra,
- Abstract summary: We break memorization down into a taxonomy: recitation of highly duplicated sequences, reconstruction of inherently predictable sequences, and recollection of sequences that are neither.
By analyzing dependencies and inspecting the weights of a predictive model, we find that different factors influence the likelihood of memorization differently depending on the taxonomic category.
- Score: 22.271015657198927
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Memorization in language models is typically treated as a homogenous phenomenon, neglecting the specifics of the memorized data. We instead model memorization as the effect of a set of complex factors that describe each sample and relate it to the model and corpus. To build intuition around these factors, we break memorization down into a taxonomy: recitation of highly duplicated sequences, reconstruction of inherently predictable sequences, and recollection of sequences that are neither. We demonstrate the usefulness of our taxonomy by using it to construct a predictive model for memorization. By analyzing dependencies and inspecting the weights of the predictive model, we find that different factors influence the likelihood of memorization differently depending on the taxonomic category.
Related papers
- A Geometric Framework for Understanding Memorization in Generative Models [11.263296715798374]
Recent work has shown that deep generative models can be capable of memorizing and reproducing training datapoints when deployed.
These findings call into question the usability of generative models, especially in light of the legal and privacy risks brought about by memorization.
We propose the manifold memorization hypothesis (MMH), a geometric framework which leverages the manifold hypothesis into a clear language in which to reason about memorization.
arXiv Detail & Related papers (2024-10-31T18:09:01Z) - Understanding Memorisation in LLMs: Dynamics, Influencing Factors, and Implications [14.818820873377303]
We study whether and to what extent large language models (LLMs) have memorised training data.
We create an experimental framework that is based on repeatedly exposing LLMs to random strings.
We identify factors that make some strings easier to memorise than others, and we identify the role of local prefixes and global context in memorisation.
arXiv Detail & Related papers (2024-07-27T14:00:21Z) - Demystifying Verbatim Memorization in Large Language Models [67.49068128909349]
Large Language Models (LLMs) frequently memorize long sequences verbatim, often with serious legal and privacy implications.
We develop a framework to study verbatim memorization in a controlled setting by continuing pre-training from Pythia checkpoints with injected sequences.
We find that (1) non-trivial amounts of repetition are necessary for verbatim memorization to happen; (2) later (and presumably better) checkpoints are more likely to memorize verbatim sequences, even for out-of-distribution sequences.
arXiv Detail & Related papers (2024-07-25T07:10:31Z) - Causal Estimation of Memorisation Profiles [58.20086589761273]
Understanding memorisation in language models has practical and societal implications.
Memorisation is the causal effect of training with an instance on the model's ability to predict that instance.
This paper proposes a new, principled, and efficient method to estimate memorisation based on the difference-in-differences design from econometrics.
arXiv Detail & Related papers (2024-06-06T17:59:09Z) - A Multi-Perspective Analysis of Memorization in Large Language Models [10.276594755936529]
Large Language Models (LLMs) show unprecedented performance in various fields.
LLMs can generate the same content used to train them.
This research comprehensively discussed memorization from various perspectives.
arXiv Detail & Related papers (2024-05-19T15:00:50Z) - Measures of Information Reflect Memorization Patterns [53.71420125627608]
We show that the diversity in the activation patterns of different neurons is reflective of model generalization and memorization.
Importantly, we discover that information organization points to the two forms of memorization, even for neural activations computed on unlabelled in-distribution examples.
arXiv Detail & Related papers (2022-10-17T20:15:24Z) - Amortized Inference for Causal Structure Learning [72.84105256353801]
Learning causal structure poses a search problem that typically involves evaluating structures using a score or independence test.
We train a variational inference model to predict the causal structure from observational/interventional data.
Our models exhibit robust generalization capabilities under substantial distribution shift.
arXiv Detail & Related papers (2022-05-25T17:37:08Z) - Counterfactual Memorization in Neural Language Models [91.8747020391287]
Modern neural language models that are widely used in various NLP tasks risk memorizing sensitive information from their training data.
An open question in previous studies of language model memorization is how to filter out "common" memorization.
We formulate a notion of counterfactual memorization which characterizes how a model's predictions change if a particular document is omitted during training.
arXiv Detail & Related papers (2021-12-24T04:20:57Z) - Deducing neighborhoods of classes from a fitted model [68.8204255655161]
In this article a new kind of interpretable machine learning method is presented.
It can help to understand the partitioning of the feature space into predicted classes in a classification model using quantile shifts.
Basically, real data points (or specific points of interest) are used and the changes of the prediction after slightly raising or decreasing specific features are observed.
arXiv Detail & Related papers (2020-09-11T16:35:53Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.