Memorization in Language Models

Overview

Researchers study when models reproduce training text or personal data, retain factual knowledge, and generalize beyond examples—questions that matter for privacy, benchmark validity, and reliable use. Common approaches include canary and text-extraction attacks, membership inference, matched training and non-training comparisons, and mechanistic analysis of activations and training dynamics; evaluations include privacy audits and unlearning benchmarks such as TOFU and MUSE. Mitigations range from data filtering and training-objective changes to fine-tuning, unlearning, and targeted edits to model activations or parameters, while agent-memory benchmarks test long-horizon recall. A central tension is that storing information does not guarantee it will be revealed, and reducing leakage must preserve useful recall and generalization.

Latest papers

126 papers12 papers in the last four weeks, level with the four weeks before. 0.1% of all new papers.

Jan 16, 2026

Spurious Rewards Paradox: Mechanistically Understanding How RLVR Activates Memorization Shortcuts in LLMs

Reinforcement Learning with Verifiable Rewards (RLVR) is highly effective for enhancing LLM reasoning, yet recent evidence shows models like Qwen 2.5 achieve significant gains even with spurious or incorrect rewards. We investigate this phenomenon and identify a "Perplexity Paradox": spurious RLVR triggers a divergence where answer-token perplexity drops while prompt-side coherence degrades, suggesting the model is bypassing reasoning in favor of memorization. Using Path Patching, Logit Lens, JSD analysis, and Neural Differential Equations, we uncover a hidden Anchor-Adapter circuit that facilitates this shortcut. We localize a Functional Anchor in the middle layers (L18-20) that triggers the retrieval of memorized solutions, followed by Structural Adapters in later layers (L21+) that transform representations to accommodate the shortcut signal. Finally, we demonstrate that scaling specific MLP keys within this circuit allows for bidirectional causal steering-artificially amplifying or suppressing contamination-driven performance. Our results provide a mechanistic roadmap for identifying and mitigating data contamination in RLVR-tuned models. Code is available at https://github.com/idwts/How-RLVR-Activates-Memorization-Shortcuts.
Jan 9, 2026

MemBuilder: Reinforcing LLMs for Long-Term Memory Construction via Attributed Dense Rewards

Maintaining consistency in long-term dialogues remains a fundamental challenge for LLMs, as standard retrieval mechanisms often fail to capture the temporal evolution of historical states. While memory-augmented frameworks offer a structured alternative, current systems rely on static prompting of closed-source models or suffer from ineffective training paradigms with sparse rewards. We introduce MemBuilder, a reinforcement learning framework that trains models to orchestrate multi-dimensional memory construction with attributed dense rewards. MemBuilder addresses two key challenges: (1) Sparse Trajectory-Level Rewards: we employ synthetic session-level question generation to provide dense intermediate rewards across extended trajectories; and (2) Multi-Dimensional Memory Attribution: we introduce contribution-aware gradient weighting that scales policy updates based on each component's downstream impact. Experimental results show that MemBuilder enables a 4B-parameter model to outperform state-of-the-art closed-source baselines, exhibiting strong generalization across long-term dialogue benchmarks.
Jul 20, 2025

Understanding the Interplay between Memorization and Learning in Large Language Models

We investigate foundational questions about the interplay between memorization and learning in large language models (LLMs). When an LLM generates a string, can we disentangle the roles of rote memorization and contextual learning? Rote memorization results in an LLM regurgitating specific details of a training string, while contextual learning leads to the generation of new strings that follow generalizable patterns of the underlying language. Related questions of interest include: can an LLM avoid memorization when optimally learning a language?, and if not, can we design training schemes to reduce memorization and improve learning? To address these questions, we propose contextual memorization, a new measure that precisely disentangles memorization from learning. Using this measure, we establish that different training strings are memorized at different epochs in training, and that memorization of some training strings is unavoidable when optimally learning a language. Finally, we propose memorization-aware training, a scheme that dynamically down-samples training strings based on their memorization, and in doing so, equalizes learning across strings, reduces memorization, and improves learning. Our conclusions are supported by extensive experiments on multiple LLMs using formal languages as a controlled testbed, enabling precise language specification, exact string sampling, and elimination of data contamination; and further validated on natural languages.
Mar 4, 2025

Evaluating Harmful Memorization in LLM Code Generation

Assessing memorization in LLM code generation is challenging because code reuse is not inherently harmful. Correct programs often share APIs, templates, and algorithmic idioms, so code similarity alone may reflect benign reuse rather than problematic memorization. We focus on the harmful case: a model fails to adapt when task semantics change and instead produces code that remains highly similar to the original solution. We propose code rewriting, an automated semantic perturbation pipeline that constructs superficially similar but semantically different variants of existing coding tasks by modifying a core logic element of the reference solution and regenerating the corresponding prompt. To summarize this behavior, we introduce the Harmful Memorization Index (HMI), an operational diagnostic that combines similarity to the original solution with performance degradation on rewritten tasks. We use harmful memorization as an operational behavioral term for this specific failure mode; HMI does not establish whether an original task or solution appeared in a model's training data. HMI is intended to complement standard accuracy metrics by capturing brittleness under small semantic changes. Across MBPP+ and BigCodeBench, we find that: (1) HMI does not monotonically increase with model scale and often decreases on simpler tasks; (2) high-HMI failures are concentrated in specific logic-edit types, especially constraint changes and data transformations; (3) supervised fine-tuning improves original-task accuracy but can substantially increase HMI on held-out tasks; and (4) across the evaluated post-training settings, PPO yields a more favorable accuracy--HMI trade-off than SFT.
Nov 11, 2024

Continual Memorization of Factoids in Language Models

As new knowledge rapidly accumulates, language models (LMs) with pretrained knowledge quickly become obsolete. A common approach to updating LMs is fine-tuning them directly on new knowledge. However, recent studies have shown that fine-tuning for memorization may be ineffective in storing knowledge or may exacerbate hallucinations. In this work, we introduce a setting we call continual memorization, where a model must memorize and retain a set of factoids through multiple stages of fine-tuning on subsequent datasets. We characterized the forgetting patterns through extensive experiments and show that LMs widely suffer from forgetting, especially when needing to memorize factoids in the second stage. We posit that forgetting can be alleviated by modifying training dynamics: (1) protecting the memorization process when learning factoids or (2) reducing interference from subsequent training stages. Intriguingly, we find that mixing randomly generated word sequences or generic data sampled from pretraining corpora at different training stages effectively mitigates forgetting REMIX: Random and Generic Data Mixing). REMIX can recover performance from severe forgetting, outperforming replay methods and other continual learning baselines. We analyze how REMIX influences the learning process and find that robust memorization follows a distinct pattern: the model stores factoids in earlier layers than usual and diversifies the layers that retain them, which results in easier recall and manipulate of the learned factoids.
Oct 3, 2024

Mitigating Memorization In Language Models

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 problematic, for example, when data are private or sensitive. In this work, we investigate methods to mitigate memorization: three regularizer-based, three finetuning-based, and eleven machine unlearning-based methods, with five of the latter being new methods that we introduce. We also introduce TinyMem, a suite of small, computationally-efficient LMs for the rapid development and evaluation of memorization-mitigation methods. We demonstrate that the mitigation methods that we develop using TinyMem can successfully be applied to production-grade LMs, and we determine via experiment that: regularizer-based mitigation methods are slow and ineffective at curbing memorization; fine-tuning-based methods are effective at curbing memorization, but overly expensive, especially for retaining higher accuracies; and unlearning-based methods are faster and more effective, allowing for the precise localization and removal of memorized information from LM weights prior to inference. We show, in particular, that our proposed unlearning method BalancedSubnet outperforms other mitigation methods at removing memorized information while preserving performance on target tasks.