Catastrophic Forgetting

Ask about this topic

Overview

Catastrophic forgetting occurs when learning new tasks or adapting to shifting data degrades performance on earlier tasks, a central challenge for systems expected to learn over time. Common approaches include replaying examples, distilling prior outputs, regularizing important weights or gradients, and isolating updates with adapters or modular experts. Researchers study these methods in sequential vision and language benchmarks as well as foundation-model fine-tuning, multimodal systems, and robotics, measuring both new-task performance and retention. No single approach works best everywhere: stronger protection can limit adaptation, while memory, compute, and benchmark choices affect results; apparent forgetting may also reflect knowledge becoming inaccessible rather than erased.

Latest papers

CardsList

351 papers59 papers in the last four weeks, up 119% on the four weeks before. 0.3% of all new papers.

Jul 30, 2026

FedOGL: Combating Catastrophic Forgetting in Federated Open-World Multimodal Graph Learning

Federated graph learning enables collaborative training over decentralized graph data without sharing raw graph information. As such risks evolve, clients must learn emerging classes from private multimodal graph streams, retain historical categories, and reject samples outside the known class space. In this setting, clients must learn emerging classes from private multimodal graph streams while preserving historical categories and rejecting samples outside the current known class space. The core challenge is catastrophic forgetting, which in federated multimodal graphs is not merely a classifier-level failure: old knowledge can be erased through modality-semantic overwriting, topology-induced structural erosion, and federated memory fragmentation. To address this challenge, we propose FedOGL, a semantic-structural memory preservation framework. On the client side, FedOGL preserves historical decision behavior through replay and task-start distillation, while protecting graph-propagation memory via projection onto a globally shared structure basis. On the server side, FedOGL maintains and transfers compact category prototypes to facilitate cross-client knowledge sharing without exposing raw graph data. Extensive experiments demonstrate that, compared with the best-performing baselines, FedOGL reduces performance degradation caused by catastrophic forgetting by 42.67%, while maintaining or improving performance on downstream tasks.

Zekai Chen, Haodong Lu, Shihao Li +5

Beijing Institute of Technology, Beijing, China

Multimodal Federated LearningFederated Continual LearningFederated Graph LearningMultimodal Graph LearningCatastrophic Forgetting
AskPDF
Jul 29, 2026

The Art of Not Forgetting A Local Learning Architecture for Continual Learning

We introduce CMP (Cognitive Memory Primitive), a continual-learning architecture that repre?sents inputs as sparse relational codes, stores them in a two-tier competitive memory, and learns through local updates without end-to-end backpropagation through its feature-generating system. We investigate whether combining sparse representations, local learning, and persistent memory can reduce catastrophic forgetting relative to conventional backpropagation-based continual?learning approaches. On a controlled domain-incremental byte-level language modeling protocol, CMP demonstrates substantially lower backward transfer than a parameter-matched Trans?former trained with online Elastic Weight Consolidation (EWC). Across a three-seed replicated 15-domain experiment, CMP exhibits stable forgetting behavior, while separate head-to-head comparisons and domain-order analyses show consistently lower forgetting than the evaluated Transformer baseline under the reported experimental settings. We report these findings alongside a substantial single-domain accuracy gap relative to the Transformer, a null result on a vision benchmark, and a documented failure to combine CMP with an independent accuracy-improving mechanism, reflecting our commitment to reporting both positive and negative outcomes. These results suggest that the combination of sparse representations, local learning, and persistent memory is a promising direction for continual learning, while motivating further investigation into the respective roles of learning rules, representations, and architectural design in mitigating catastrophic forgetting.

Ashmith Atmuri, Yashaswini Rao Bhogarajula

Arkadhi Research

Memory-Augmented Neural NetworksContinual LearningDomain-Incremental LearningIncremental LearningLocal LearningCatastrophic Forgetting
AskPDF
Jul 29, 2026

ForgetBench: Benchmarking Forgetting Dynamics of Long-Term Parametric Memory in Language Models

Large language models (LLMs) have demonstrated strong capabilities in knowledge acquisition and reasoning, yet their ability to retain previously acquired knowledge under repeated updates remains insufficiently understood. Existing evaluation paradigms primarily focus on single-step reasoning or static knowledge editing, which fail to capture the temporal dynamics of knowledge retention and degradation during continual model modification. In this work, we propose ForgetBench, a benchmark for measuring forgetting dynamics in the long-term parametric memory of LLMs under continual knowledge editing. ForgetBench introduces two complementary evaluation paradigms, namely concept-based QA and scenario-based QA, to probe the retention of isolated facts and structured relational knowledge. Building upon a sequential editing framework, we construct temporally ordered knowledge streams and evaluate model behavior across multiple editing stages. To quantify these dynamics, we introduce a unified evaluation framework that indexes memory age by the number of subsequent edits and measures temporal decay, retention strength, and cross-instance stability. Extensive experiments across diverse models and editing methods reveal distinct forgetting trajectories and a recurring tension between long-term retention and generalization quality. Our findings highlight the need for more robust memory mechanisms that can effectively acquire, update, and preserve knowledge over time in future LLMs.

Ruxi Gu, Zhenliang Zhang, Zilei Wang +1

Department of Automation, University of Science and Technology of China, Hefei, China · State Key Laboratory of General Artificial Intelligence, BIGAI, Beijing, China · School of Computer and Communication Engineering, University of Science and Technology Beijing, Beijing, China.

Knowledge EditingContinual Learning for LLMsLLM EvaluationPersistent Memory for Language ModelsLLM MemoryCatastrophic Forgetting
AskPDF
Jul 26, 2026

The Intruder Threshold: A Spectral Law for LoRA Fine-Tuning

LoRA fine-tuning can create intruder dimensions: new leading singular vectors of the updated weight matrix W+BAW+BA that are nearly orthogonal to all pretrained singular vectors and that drive catastrophic forgetting. Since their discovery, no theory has predicted, layer by layer on measured spectra, when they appear. We derive a per-layer critical update strength s∗=θˉ/(γσ1(BA))s^\ast=\barθ/(γσ_1(BA)), computed from the measured spectrum of WW alone through the rectangular spiked-deformation transform, together with an exact secular-equation characterization of the updated spectrum, with no fitted parameters. In a pre-specified study spanning four dense Transformer families, a state-space model, a mixture-of-experts model, and an encoder-decoder (18 adapters, 9{,}840 layer scans), the law localizes the empirical threshold within a factor of two on 82%82\% of layers, separates intruder-bearing from intruder-free layers at deployment with a mean AUC of 0.890.89, holds unchanged on six third-party adapters, and predicts where WikiText-2 perplexity begins to degrade; a combination of the two pre-specified edge evaluations reaches 98%98\% and is confirmed out-of-bag on the external adapters (0.9970.997). Full fine-tuning disperses its update far below the threshold of every layer, which resolves the asymmetry between LoRA and full fine-tuning. Norm-matched interventions confirm that threshold-crossing layers, rather than update magnitude, carry the forgetting, and a spike-budget rule derived from the thresholds, requiring one SVD and no validation sweeps, reduces forgetting by 62%62\% on the most fragile model at no task cost.

Peng Xie

Technical University of Munich, Campus Heilbronn

LLM Fine-TuningLow-Rank AdaptationCatastrophic Forgetting
AskPDF
Jul 24, 2026

NeuroSynth: A Biologically Inspired Continual Reinforcement Learning Architecture for Mitigating Catastrophic Forgetting

Artificial Intelligence (AI) systems often perform well on isolated tasks but struggle under continual learning conditions, where training on new tasks can overwrite previously acquired knowledge, a failure mode known as catastrophic forgetting. Biological learning systems reduce this interference through complementary memory processes involving rapid hippocampal encoding and slower cortical consolidation. This study introduces NeuroSynth, a brain-inspired continual reinforcement learning architecture designed to mitigate catastrophic forgetting through a dual-pathway consolidation mechanism. NeuroSynth separates rapid task acquisition from long-term retention using distinct "plan" and "habit" pathways combined with replay and knowledge distillation. NeuroSynth was evaluated against Proximal Policy Optimization (PPO) and Elastic Weight Consolidation (EWC) across three sequential navigation tasks with changing goal locations in a non-revisitation continual learning setting. Across six independent seeds, NeuroSynth preserved substantially more early-task knowledge than PPO after sequential training, achieving 18.00% Task A success rate compared to 0.33% for PPO (p = 0.014929, Cohen's d = 1.49) and 35.33% Task B success rate compared to 0.00% for PPO (p = 0.002376, Cohen's d = 2.31). NeuroSynth also demonstrated higher final Task C performance than EWC, achieving 9.00% compared to 2.00% (p = 0.226643, Cohen's d = 0.56), indicating a moderate but not statistically significant advantage. These findings suggest that biologically inspired consolidation mechanisms may improve the stability-plasticity balance in continual reinforcement learning systems.

Yash Kini

James Madison High School Vienna, Virginia, USA

Continual LearningContinual RLCatastrophic Forgetting
AskPDF
Jul 23, 2026

Source-Prior-Driven Selective Adaptation for Efficient Diffusion Model Finetuning

Fine-tuning large diffusion models for new domains or styles involves a trade-off: improving target-specific generation often degrades the pretrained model's broad generative capability. Existing full and parameter-efficient fine-tuning methods typically handle this trade-off only implicitly. In this work, we propose a novel source-prior-driven selective adaptation method to efficiently fine-tune diffusion models, achieving a favorable trade-off. Our method relies on two key observations: (1) the loss of general generative capability is highly inconsistent across pretrained parameters, and (2) parameters that have a relatively small impact on the model's general generative capability remain structurally inconsistent across layers and parameter types. Motivated by these observations, we first learn a static mask to explicitly identify parameters better suited for downstream adaptation, and then construct structured update strategies for the selected subset. Experiments show that our method achieves a better adaptation-retention trade-off than existing strong baselines.

Yi Xiong, Yuan-Yuan Cheng, Xiao-Ming Fu

University of Science and Technology of China, China

Fine-TuningDiffusion Model Fine-TuningSelective Fine-TuningParameter-Efficient Fine-TuningCatastrophic ForgettingDiffusion Model Adaptation
AskPDF
Jul 22, 2026

The World Model Remembers, the Actor Forgets: Dream Rehearsal for Continual Model-Based RL

Model-based reinforcement-learning agents of the DreamerV3 family forget catastrophically when trained on task sequences, even when an unbounded replay buffer preserves every earlier experience. We ask a question the continual-RL literature has assumed an answer to but never measured: which component forgets? Under never-clear replay, pre-registered component-level probes (n=3 seeds throughout) show that the world model retains essentially everything measurable about old tasks -- reward discrimination (retention ratio ~1.0), value estimates, and termination structure -- while the actor's behavior collapses. Forgetting in this regime is a channel problem, not a memory problem. We demonstrate this by intervention: with the world model frozen and identical imagined rollouts, reinforcement learning in imagination fails to recover a lost skill (0/3 seeds), while supervised self-imitation on the world model's own graded dreams recovers it on 3/3 seeds with zero environment interaction. Interleaved during training, this graded dream rehearsal yields a task-label-free, parameter-constant continual learner: 3/3 four-task chains retained where plain replay passes 0/3, 3/3 eight-task chains, and consistent gains over matched real-episode cloning (paired difference +0.13, bootstrap 95% CI [0.07, 0.24], complete seed separation). The dream-grading step is load-bearing: we characterize two scoring failure modes, provide an offline selection gauge that caught both before they contaminated results, and give a realized-first grading rule that closes them. All experiments were pre-registered with committed protocols; every refuted hypothesis is reported.

Gurp Nijjer

Quantegra Research, Canada

Generative ReplayReinforcement LearningReplay-Based Continual LearningContinual RLModel-Based RLCatastrophic Forgetting
AskPDF
Jul 21, 2026

Code Division Modulation Layers Against Forgetting and Inference in Continual Gait Identification

Continual learning (CL) has been recently employed in biometric identification systems thanks to its ability to integrate new knowledge within a pre-trained model and to the possibility of reducing the computational cost of training. Unfortunately, such approaches pose new challenges both in terms of final accuracy and privacy guarantees since a progressive fine-tuning of the model on small subsets expose them to catastrophic forgetting and successful inference attacks. This paper evaluates the efficiency of code division modulation layers (CDML) on a gait identification system which has been trained following a continual learning policy. The proposed approach preserves accuracy on all the tasks while mitigating membership inference attacks at the same time. Moreover, the impact of retransmission is minimized since replaying data is not necessary.

Simone Milani

Department of Information Engineering, University of Padova, Padova, Italy

Membership Inference AttacksPrivacy-Preserving MLGait RecognitionCatastrophic Forgetting
AskPDF
Jul 20, 2026

The Art of Not Forgetting

We introduce CMP (Cognitive Memory Primitive), an architecture that represents inputs as sparse relational codes, stores them in a two-tier competitive memory, and learns entirely through local, gradient-free updates, with no backpropagation anywhere in the network. We use this architecture to test a specific hypothesis: that catastrophic forgetting, usually treated as a training-time defect to be patched with replay or regularization, is instead a structural consequence of how backpropagation assigns credit and that a learning rule that is local and sparse by construction should resist it without a patch. On a controlled domain-incremental protocol across 15 text domains, three-seed replicated, CMP's backward transfer is 15-19x better than a matched-size Transformer trained with online EWC, and the result survives a domain-order control (reported as a range, +0.24 to +0.44, rather than a single figure). We report this alongside a real, substantial accuracy gap versus the Transformer baseline, a null result on a recognized vision benchmark, and a diagnosed, unresolved failure attempting to combine this architecture with a separate mechanism that improves raw accuracy, disclosed because an honest negative result is more useful than an omitted one. The central claim is narrow and falsifiable: local, sparse, non-backpropagation learning measurably resists catastrophic forgetting better than backpropagation with its standard fix, under conditions we state precisely.

Ashmith Atmuri, Akshay Kumar, Yashaswini Rao Bhogarajula

Arkadhi Labs

Memory-Augmented Neural NetworksDomain-Incremental LearningIncremental LearningLocal LearningCatastrophic Forgetting
AskPDF
Jul 17, 2026

Rethinking Quantum Continual Learning with Quantum Fisher Information

Quantum continual learning aims to train quantum models on sequential tasks without losing previously learned knowledge. However, variational quantum classifiers (VQCs) are prone to catastrophic forgetting under nonstationary task distributions. We propose quantum elastic weight consolidation (QEWC), a quantum Fisher information (QFI)-informed regularization method for mitigating forgetting. Unlike conventional elastic weight consolidation based on classical Fisher information (CFI), which measures parameter importance through measurement-dependent output statistics, QEWC uses QFI to quantify the intrinsic sensitivity of the parameterized quantum state. This gives an information-geometric view in which important parameters are identified by the local response of the quantum state manifold. We evaluate QEWC on VQCs trained on sequential binary classification tasks, including classical image-classification and quantum phase-classification tasks. Simulations show that sequential training without regularization causes severe forgetting, while both CFI-based EWC and QFI-based QEWC improve retention of previous tasks. Mechanistic analyses further show that the two methods impose different regularization geometries: CFI acts selectively on measurement-sensitive directions, whereas QFI imposes a denser state-geometric constraint over parameter space. Under depolarizing noise, CFI values are strongly suppressed by degraded measurement statistics, while QFI preserves a more stable sensitivity structure of the noisy parameterized quantum state. These results establish QEWC as a physically motivated approach for studying and mitigating forgetting in quantum continual learning through quantum-state geometry.

Yu-Chao Hsu, Yu-Cheng Lin, Tai-Yue Li +2

School of Electrical Engineering, Korea Advanced Institute of Science and Technology (KAIST), 291 Daehak-ro, Yuseong-gu, Daejeon 34141, Republic of Korea · Department of Electrophysics, National Yang Ming Chiao Tung University, Hsinchu 300092, Taiwan. · National Center for High-Performance Computing , National Institutes of Applied Research (NIAR), Hsinchu 300092, Taiwan

Quantum Machine LearningVariational Quantum CircuitsFisher InformationCatastrophic Forgetting
AskPDF
Jul 16, 2026

Towards Human-like Physical Intelligence: LifelongVision-Language-Action Learning for Robotic Manipulation

Similar to the natural capabilities of humans to sequentially learn new tasks, robots with Vision-Language-Action (VLA) models should possess lifelong learning ability to learn a new task when deployed in open-world environments. However, most recently proposed lifelong learning models aim to effectively learn the current task (plasticity) or maintain high accuracy on previous tasks (stability), while the plasticity-stability trade-off remains largely unsolved in robotic manipulation models. To address this fundamental challenge, we propose a cache-efficient lifelong Vision-Language-Action learning framework for robotic manipulation (i.e., LifelongVLA), which alleviates the plasticity-stability trade-off with a dual-timescale adaptation mechanism while achieving low-cost robotic deployment with a cache-efficient replay strategy. More concretely, we propose a dual-timescale LoRA gating module to decompose VLA adaptation into two lightweight pathways: a short-term adapter for plasticity and a long-term adapter for stable consolidation. These pathways are integrated via a task-aware gate, enabling explicit control of the plasticity-stability trade-off. In the skill replay phase, a cache-efficient stochastic replay strategy is proposed to preserve more balanced retention signals without full-trajectory storage. Finally, experiments show that LifelongVLA outperforms existing baselines, demonstrating efficient skill expansion, robust retention of learned manipulation behaviors, and reduced reliance on retraining for real-world deployment on an xArm robot.

Yao He, Gan Sun, Wenqi Liang +2

South China University of Technology · University of Trento

Continual Robot LearningRobot Skill LearningLow-Rank AdaptationContinual Learning for VLA ModelsRobotic ManipulationVision-Language-Action Models+1
AskPDF
Jul 15, 2026

Symbiosis-Inspired Knowledge Distillation for Incremental Object Detection

Incremental object detection (IOD) aims to extend detectors to new categories while retaining previously acquired knowledge. Existing methods often adopt a class incremental learning perspective, separating feature spaces to sharpen decision boundaries. However, this separation-oriented paradigm may overlook object symbiosis in detection, where co-occurrence and occlusion introduce spatial and semantic dependencies that benefit from shared representations. Ignoring these dependencies distorts the shared representations, exacerbates confusion between old and new classes, and accelerates catastrophic forgetting. To address this, we propose Symbiosis-Inspired Knowledge Distillation (SIKD), which explicitly leverages object symbiosis at two complementary levels. Spatial Symbiosis Distillation (SpSD) focuses on symbiotic regions where the old model responds with high overlap to objects in the new task. It preserves generalizable old class cues, suppresses class-specific bias and redundancy, and distills the refined evidence to the new model at matched spatial locations with slot-aligned supervision. Semantic Symbiosis Distillation (SeSD) maintains class level structure by forming confidence weighted prototypes for old classes and aligning their inter class soft ranks over the old class logits, which stabilizes the semantic topology during adaptation. Extensive experiments demonstrate the effectiveness and superiority of the proposed method.

Mingyue Zeng, De Cheng, Zhipeng Xu +3

State Key Laboratory of Integrated Services Networks, School of Telecommunications Engineering, Xidian University, Xi’an, China · School of Electronic Engineering, Xidian University, Xi’an, China

Class-Incremental LearningObject DetectionKnowledge DistillationCatastrophic Forgetting
AskPDF
Jul 14, 2026

Domain-Incremental Remote Sensing Change Detection via Difference-Guided Adaptation and Frequency-Decoupled Distillation

Remote sensing change detection (RSCD) models are prone to catastrophic forgetting when incrementally adapted to new domains. Existing domain-incremental learning (DIL) methods mainly preserve image-level representations but often overlook bitemporal discrepancy cues, which are critical for robust change detection under domain shifts. To address this limitation, we propose DG-FDD, a domain-incremental change detection framework that integrates Difference-Guided Adaptation and Frequency-Decoupled Distillation. Specifically, the Difference-Guided Dynamic Adapter (DGDA) models bitemporal feature discrepancies to promote change-aware feature adaptation and reduce domain-specific interference. Meanwhile, the Frequency-Decoupled Knowledge Distillation strategy with Cross-domain Synthesis (FDKD-CS) separates structural information from domain style in the frequency domain, enabling stable knowledge transfer without historical data. Extensive experiments on three public high-resolution RSCD datasets under two- and three-domain incremental protocols demonstrate that DG-FDD effectively mitigates catastrophic forgetting. Compared with independently trained single-task models, DG-FDD records mean relative changes in F1 and IoU of only -0.23% and -0.45%, respectively, across six two-domain sequences, and -0.69% and -1.31%, respectively, across the three evaluated three-domain sequences. These results indicate a favorable stability-plasticity balance between historical knowledge retention and new-domain adaptation in continual cross-domain change detection.

Daifeng Peng, Yaning Li, Haiyan Guan

School of Remote Sensing and Geomatics Engineering, Nanjing University of Information Science and Technology, Nanjing 210044, China

Remote Sensing Change DetectionDomain-Incremental LearningKnowledge DistillationCatastrophic Forgetting
AskPDF
Jul 14, 2026

Traceback Translators Against Forgetting in Continual Fake Speech Detection

Fake speech detectors are increasingly challenged by the development of new and more accurate generative models. To cope with this problem, continual learning techniques are nowadays widely considered feasible strategies for updating models to new datasets, but they also lead to decreased performance on previously seen samples (catastrophic forgetting). In this work, we propose a forgetting-resilient solution based on the adoption of domain translators within a frozen detector, which remaps the new feature spaces into the original ones by means of a traceback translator network. Experimental results show that this strategy enables the achievement of high detection rates with respect to traditional retraining, while minimizing the computational effort and preserving the detection accuracy on previous data.

Enrico Gottardis, Mattia Tamiazzo, Simone Milani

Department of Information Engineering · University of Padova · Padova, Italy

Audio Deepfake DetectionCatastrophic Forgetting
AskPDF
Jul 13, 2026

An Empirical Analysis of Continual Learning for Heterogeneous Medical Visual Question Answering

Deploying medical visual question answering (MedVQA) systems in real-world clinical settings requires models that adapt to new clinical tasks without forgetting previously acquired knowledge. Continual learning (CL) provides a practical framework for this setting. Despite rapid progress in medical vision-language models, the behavior of CL methods when training these models across heterogeneous MedVQA tasks remains underexplored. This work presents a systematic evaluation of CL for MedVQA across diverse clinical objectives, including classification, multi-label classification, detection, cell counting, and report generation. Specifically, we explore (1) the ability of existing CL methods to mitigate catastrophic forgetting; (2) their sensitivity to task ordering, analyzing how different task sequences influence performance retention and forgetting; and (3) the evolution of low-rank adaptation parameters as new tasks are learned, revealing patterns of weight drift under different CL methods. Our findings suggest that existing CL methods struggle to maintain stability-plasticity balance when tasks with different objectives and supervision formats are interleaved. Code and full experimental setup will be publicly available.

Mai A. Shaaban, Tausifa Jan Saleem, Alaa Mohamed +3

Mohamed bin Zayed University of Artificial Intelligence, Abu Dhabi, UAE · Department of Mathematics and Computer Science, Faculty of Science, Alexandria University, Alexandria, Egypt

Medical VQAContinual Learning for VLMsCatastrophic Forgetting
AskPDF
Jul 13, 2026

Unified Gradient Projection: Language-Balanced Continual Learning for Multilingual Low-Resource ASR

Large-scale pretrained ASR models such as Whisper exhibit strong multilingual capabilities. However, fine-tuning on low-resource languages often causes catastrophic forgetting. Although continual learning mitigates this issue, existing methods struggle to regulate cross-task interference in multilingual settings, where dominant languages bias optimization. We propose Unified Gradient Projection (UGP), which constrains parameter updates using reference gradients from language-balanced replay in a unified projection space. By equalizing per-language contributions in the projection, UGP reduces dominant-language bias and improves cross-lingual stability. We further show that combining gradient-level projection with data-level replay yields complementary gains in stability and plasticity. Across diverse low-resource language groups and model scales, UGP enables effective adaptation while substantially mitigating forgetting. On Whisper-large-v3, it achieves near-zero average forgetting.

Ziang Ren, Guodong Lin, Yuchen Ai +2

Department of Electronic Engineering, Tsinghua University, China

Automatic Speech RecognitionGradient ProjectionLow-Resource Speech RecognitionCatastrophic ForgettingMultilingual ASR
AskPDF
Jul 13, 2026

CA-DGCL: Dynamic Graph Continual Learning via Condensation and Attachment

Dynamic graph continual learning (DGCL) is an effective manner for handling catastrophic forgetting in dynamic graphs. However, existing DGCL methods underutilize temporal information across graph snapshots. To address this critical issue, we propose a novel framework for Dynamic Graph Continual Learning via Condensation and Attachment (CA-DGCL). Specifically, CA-DGCL first condenses historical graph snapshots into compact semantic representations efficiently. Further, a cross-timestamp node chains is built to construct a third-order tensor and Tucker decomposition is applied to this tensor for obtaining stable node features, which encapsulate historical knowledge. Finally, these node features are used to generate new nodes and attached to the current graph for replaying of past information without compromising the new patterns. In addtion, a refined forgetting measure is introduced to make it more suitable for dynamic graph settings. Extensive experiments demonstrate that CA-DGCL outperforms baselines in forgetting suppression as well as maintain competitive accuracy, proving its efficacy for dynamic graph continual learning.

Tingxu Yan Ye Yuan

1Tingxu Yan and Ye Yuan* are with the College of Computer and Information Science, Southwest University, Chongqing 400715, China

Replay-Based Continual LearningCatastrophic Forgetting
AskPDF
Jul 13, 2026

Can a Language Model Learn Facts Continually in Its Weights?

Continual learning is a long-standing capability gap between LLMs and humans. Writing new knowledge into a model's weights routinely causes it to forget old knowledge, commonly denoted as "catastrophic forgetting". Various modifications of supervised fine-tuning and distillation aim to mitigate catastrophic forgetting, but quantifying what (or how much) information was forgotten is often difficult. In this paper, we study whether current methods of writing knowledge into weights enable models to learn continually without forgetting. We introduce a framework for studying continual learning in the iterative regime, writing invented facts one at a time into a Qwen3 model already modified by previous writes, and varying the training data, method, and parameter update. Across SFT and off- and on-policy distillation, using LoRA or full fine-tuning, we compare repeated statement training (the same fact repeated in two formats) with varied example training (24 factual restatements) and find that varied examples comprehensively support more flexible use. After twenty sequential writes and merges, the model answers only 1% of questions about earlier facts correctly when every write uses repeated statements, compared with 46% when every write uses varied examples. We additionally show that this retention depends on the data used for the later writes, regardless of training method or parameter update, and that behavioral forgetting of an earlier fact does not erase its presence from the log-probabilities. Together, our framework neatly provides a comparison of performance across training data, training regimes, and parameter update schemes in an iterative learning task.

Charles O'Neill, Max Kirkby, Jonathon Liu +1

Baseten

Knowledge EditingContinual Learning for LLMsFactual Knowledge in Language ModelsIncremental LearningCatastrophic Forgetting
AskPDF
Jul 10, 2026

Interference and Retention in Continual Learning

Continual learning commonly relies on post-hoc mechanisms such as replay, elastic regularization, or distillation. This work argues that forgetting should instead be modeled directly as interference between tasks. In the frozen-feature regime, forgetting from learning a new task is exactly the interference energy induced on the old task. In deep networks, the same quantity is recovered through path-averaged curvature with minimal additional forward passes. When task supports are disjoint, forgetting can be eliminated structurally and when task supports overlap in conflicting directions, a non-zero distortion floor is unavoidable. The same geometry optimally merges models through task-aware orthogonalization. From this analysis we derive Interference-Gated Functional Allocation (IGFA), a replay-free, Fisher-free method that shares directions when tasks align and protects them when they conflict. Across benchmarks, IGFA achieves lossless retention when tasks are structurally separable and moves unavoidable cost from irreversible forgetting into deferred but recoverable plasticity when they are not. It matches the strongest replay-free structural baselines on dissimilar-task streams and improves on unconditional projection when similarity makes transfer worth preserving.

Julius Störk

VARTA Microbattery GmbH, Ellwangen, Germany

Continual LearningOnline Continual LearningTask InterferenceCatastrophic Forgetting
AskPDF
Jul 8, 2026

Lifelong Representations: A Survey on Continual Self-Supervised Learning for Vision Models

Traditionally, continual learning has assumed access to labeled data, yet many real-world applications -- such as lifelong robotics -- require models to adapt continuously from unlabeled streams. This has led to the development of continual self-supervised learning (CSSL), a rapidly growing area that lacks a dedicated, systematic review. In this work, we present a comprehensive survey of CSSL for vision, with connections to emerging vision-language settings. First, we analyze existing evaluation protocols and highlight inconsistencies that hinder fair comparison. We then examine why self-supervised objectives exhibit improved robustness to catastrophic forgetting, relating this to task-agnostic representations and smoother loss landscapes. Next, we organize existing methods into a unified taxonomy based on their forgetting-mitigation strategies, including distillation, replay, regularization, architectural approaches, model merging, and objective-level adaptation. Finally, we identify open challenges such as scalability and the need for fast adaptability. We argue that advancing CSSL requires moving beyond small-scale benchmarks towards continual pre-training paradigms for large-scale systems.

Sergi Masip, Alicja Dobrzeniecka, Jonathan Swinnen +4

PSI, KU Leuven, Leuven, Belgium · NASK - National Research Institute, Warsaw, Poland · Computer Vision Center, Universitat Autonoma de Barcelona, Barcelona, Spain +1

Self-Supervised LearningSelf-Supervised Visual Representation LearningCatastrophic Forgetting
AskPDF
Jul 8, 2026

Intrinsic-Noise Consolidation: A Doob-Barrier-Conditioned Diffusion Turns Analog Device Noise into a Continual-Learning Resource

On analog neuromorphic hardware, intrinsic device noise is normally an accuracy tax. We ask whether it can instead consolidate memories. We cast per-synapse consolidation as a Doob h-transform: condition each weight's stochastic dynamics on never crossing a memory-critical barrier around its consolidated value. The conditioned diffusion gains an extra drift sigma^2 d/dw log h, a restoring force amplified by the noise variance itself that diverges at the barrier. We are explicit about novelty: the anchored drift -s(w-mu) our rule also contains is not ours (the limit of OUA, MESU, and EWC), and we surrender it. We claim only the conjunction of (a) the Doob barrier-conditioning as a synaptic rule, to our knowledge unclaimed (every h-transform use we found is generative modeling, none synaptic), and (b) a falsifiable prediction: increasing intrinsic noise non-monotonically improves sequential-task retention, an inverted-U that anchored-drift methods cannot produce. We pre-registered this as a go/no-go gate; it passes. On single-head Split-MNIST (8 seeds) the rule lifts retention 10.9 points at an interior optimum (paired Wilcoxon p=0.004), while matched OU/EWC/MESU anchors are monotone. Ablating the conditioning removes the effect; the optimum tracks the barrier; the inverted-U survives a second task stream and the realization where noise enters the forward pass. We then measure the intrinsic noise on real BrainScaleS-2 silicon (additive, trial-to-trial independent, tunable via on-chip averaging) and run the rule on the chip with its noise in the training loop: barrier-conditioning retains a prior task 15.6 points better than the matched control at matched average accuracy, a stability-plasticity shift, not a net-accuracy win (single seed; retention measured, energy modelled). Intrinsic analog noise thus becomes a consolidation dividend a digital accelerator must spend energy to generate.

Gunner Levi Howe

Continual LearningStability-Plasticity DilemmaDoob H-TransformCatastrophic ForgettingNeuromorphic ComputingMemory Consolidation+1
AskPDF
Jul 6, 2026

REDDIT: Forgetting-Resistant Correction of Timestamp Drift in ASR via Replay-Based Distribution Editing

Modern autoregressive ASR systems can emit timestamps as decoded tokens, enabling timestamped transcription without frame-level aligners or inference-time post-processing. We show that these generated timestamps can drift across long non-speech spans: the transcript may remain plausible, but the decoded time axis drifts away from the audio. We study this non-speech-induced timestamp drift with self-built gap and long-gap benchmarks across 15 evaluated timestamp-producing ASR and audio-language systems. Naive timestamp-corrected fine-tuning improves alignment but can severely degrade non-target ASR behavior, exposing a forgetting problem. We propose REDDIT(REplay-based Distribution eDITing), a lightweight two-stage post-training framework that corrects timestamps while avoiding this catastrophic forgetting: it first edits timestamp targets under the model's own replayed decoder context while matching the frozen base distribution on non-timestamp tokens, then applies a short edited-prefix refinement stage. In this framework, we construct correction supervision without human transcripts or human timestamp annotations by combining VAD-trimmed speech spans with inserted non-speech gaps and known concatenation offsets. On Whisper-tiny, 34.9 hours of targeted correction audio used and only 1.6% of model parameters updated, raising long-gap mIoU from 38.7% to 95.0% and reducing mixed-gap out-of-domain AAS from 2752 ms to 223 ms while preserving CV-en MER at 41.3% (versus 524.2% for ordinary SFT decoder tuning).

Cheng-Kang Chou, Ming-To Chuang, Ke-Han Lu +2

National Taiwan University · Carnegie Mellon University · NTU Artificial Intelligence Center of Research Excellence (NTU AI-CoRE)

ASR EvaluationSpeech ProcessingAutomatic Speech RecognitionCatastrophic Forgetting
AskPDF
Jul 5, 2026

RL Forgets! Towards Continual Policy Optimization

Continual post-training is becoming a central paradigm for adapting vision-language models to evolving tasks. Recent studies report that reinforcement learning is less prone to forgetting than supervised fine-tuning, motivating the view that RL is inherently resistant to forgetting. However, this view remains insufficiently validated, as existing evidence is largely drawn from outdated or homogeneous benchmarks. We revisit this assumption by introducing MRCL, a Multimodal Reasoning Continual Learning benchmark built from recent and diverse multimodal reasoning tasks. Experiments on MRCL show that standard reinforcement learning still suffers from catastrophic forgetting during continual post-training. We trace the failure to an objective mismatch. The KL regularization used in common policy optimization methods is evaluated on current-task data, whereas forgetting is caused by behavioral drift on prior-task distributions. To address this problem, we propose Continual Policy Optimization (CPO), a replay-free method grounded in a prior-task behavioral KL objective. CPO derives a local Fisher surrogate from the historical KL objective and uses parameter movement as a gradient-free proxy for Fisher sensitivity, enabling sparse regularization with negligible additional overhead. Experiments on three model scales and comparisons with multiple RL baselines show that CPO consistently reduces forgetting while maintaining effective adaptation and preserving broader pretrained capabilities. The implementation code is available at https://github.com/MaolinLuo/CPO.

Mao-Lin Luo, Zhe-Xu Wang, Zi-Hao Zhou +4

School of Computer Science and Engineering, Southeast University, Nanjing 210096, China · Key Laboratory of Computer Network and Information Integration (Southeast University), Ministry of Education, China · Zhongguancun Academy +1

RL Post-TrainingContinual Learning for VLMsContinual RLCatastrophic Forgetting
AskPDF
Jul 2, 2026

Denser ≠\neq Better: Limits of On-Policy Self-Distillation for Continual Post-Training

Continual post-training enables foundation models to acquire new knowledge while preserving existing capabilities. Recent work suggests that on-policy learning can mitigate forgetting, with on-policy self-distillation emerging as a particularly attractive approach. In this work, we revisit this optimistic view through self-distillation policy optimization (SDPO). Our experiments show that SDPO can accelerate in-domain specialization when teacher signals are stable and well aligned, but it struggles to generalize to out-of-distribution scenarios. In continual post-training, SDPO exhibits stronger forgetting and can even collapse, whereas on-policy reinforcement learning methods such as GRPO adapt more conservatively and better preserve prior capabilities. Further analyses reveal that denser self-distillation induces larger drift in both parameter space and response space, and can amplify high-frequency formatting artifacts through a self-reinforcing teacher--student loop. These findings suggest that on-policy data alone is insufficient for continual learning. Dense self-distillation can accelerate specialization when teacher targets are stable and token-level supervision is reliable, but it should not be treated as a default stabilizer for continual post-training. Our code is available at https://github.com/Moenupa/SDPO-CL.

Meng Wang, Haohan Zhao, Wenzhuo Liu +7

Centre for Artificial Intelligence and Robotics, HKISI, CAS · Institute of Automation, CAS · University of Chinese Academy of Sciences +1

Continual Learning for LLMsOn-Policy Self-DistillationRL Post-TrainingLLM Post-TrainingCatastrophic Forgetting
AskPDF
Jul 1, 2026

Rosetta: Composable Native Multimodal Pretraining

Achieving true artificial general intelligence requires foundation models capable of integrating new modalities without forgetting prior knowledge. However, accommodating continuous generative objectives alongside discrete understanding tasks causes severe gradient conflicts. Existing architectures, including standard Mixture-of-Experts (MoE), are highly susceptible to representation overwriting. Even structurally partitioned paradigms like Mixture-of-Transformers (MoT) remain vulnerable to catastrophic forgetting, severely impeding multimodal scalability. In this work, we introduce Rosetta, a composable native multimodal pretraining framework designed for seamless and non-destructive modality expansion. Rosetta adopts a modular paradigm where core foundational knowledge is preserved within global shared experts, while modality-specific capabilities are distributed across plug-and-play experts. To guarantee non-destructive composition, we propose Momentum-Anchored Orthogonal Projection (MAOP). MAOP leverages the optimizer's momentum state as an implicit semantic anchor, selectively neutralizing conflicting gradient components from new modalities while preserving synergistic updates. Extensive evaluations demonstrate that, while standard MoE and MoT architectures suffer catastrophic forgetting of previously acquired knowledge, Rosetta robustly preserves established language and visual understanding. Furthermore, it delivers superior image generation and unlocks cross-modal synergy, paving the way for truly composable and unified multimodal foundation models. To facilitate further multimodal research, we release our code and checkpoints to the community. Project page at https://rosetta-lmm.github.io/.

Xiangyue Liu, Zijian Zhang, Miles Yang +3

HKUST · Tencent Hunyuan

Unified Multimodal ModelsMultimodal PretrainingMultimodal GenerationGradient ProjectionMultimodal Foundation ModelsContinual Learning for VLMs+2
AskPDF
Jun 29, 2026

Neural Subspace Reallocation: Continual Learning as Retrieval-Based Subspace Memory Management

We introduce Neural Subspace Reallocation (NSR), which reframes continual learning as memory management over parameter subspaces. Instead of treating Low-Rank Adaptation (LoRA) modules as disposable per-task adapters, NSR manages them as compressible, retrievable memory units on a frozen backbone through a recurring cycle: (1) compress learned LoRAs via SVD, (2) reserve them in a TaskKnowledgeBank, (3) recall related past LoRAs by embedding similarity to warm-start new or returning tasks, and (4) reallocate the active subspace accordingly, with distillation protecting prior tasks. We prove that in cyclic environments any memoryless allocation policy incurs cumulative regret Omega(T(M-1)Delta_switch) relative to a history-aware policy backed by the Bank (Theorem 1). Empirically, on Split-CIFAR-100 the Bank reduces cyclic recovery time by 10x, exactly as predicted, and on the heterogeneous 5-Datasets benchmark NSR achieves the highest accuracy and the least forgetting, about 9x closer to zero backward transfer than the memoryless heuristics. Crucially, we run a controlled study that isolates which component matters: holding the Bank fixed and varying only the allocation rule, we find that a simple similarity-based retrieval rule matches or beats a learned reinforcement-learning controller (recovering recurring tasks in 0 vs 1.8 steps and reaching equal accuracy). Our central, honest finding is therefore that the memory mechanism -- compression and similarity retrieval -- rather than a learned allocation policy, drives continual-learning performance under fixed capacity. A memory-budget analysis confirms the compressed Bank stays small -- 0.29 MB of parameter memory per task -- so a top-K retention cap bounds the total footprint while preserving fast recovery for retained tasks.

Byeong Hoon Yoon

Independent Researcher

Storage-Efficient Continual LearningContinual LearningCatastrophic Forgetting
AskPDF
Jun 26, 2026

Continual Learning for Sequential Personalization of Small Language Models: A Stability Monitoring Analysis

Small Language Models (SLMs) are increasingly being considered for deployment on edge devices such as laptops, enabling private, low-latency, and locally personalized applications. However, personalization requires models to adapt over time to evolving user- or task-specific data, placing them in a continual learning setting. This creates the risk of catastrophic forgetting, where learning new information degrades performance on previously learned tasks or broader model capabilities. Recent benchmarks such as TRACE have shown that continual fine-tuning can significantly degrade the general abilities of aligned large language models. In this work, we present a study for sequential LoRA personalization of SLMs. We save model checkpoints after each adaptation stage and evaluate them on current tasks, previously seen tasks, and a fixed reference set. This checkpoint-level protocol enables us to monitor task performance, forgetting, and reference set drift over time. We show that lightweight reference set distributional diagnostics can reveal model-specific instability patterns during sequential LoRA personalization of SLMs, including cases where task-level metrics alone hide harmful adaptation. We hope this can highlight new research avenues for monitoring stability of SLMs in a continual learning setting.

Thomas S. Paula, Lucas S. Kupssinskü, Rodrigo C. Barros

MALTA, Machine Learning Theory and Applications Lab, PUCRS, Porto Alegre, Brazil · Kunumi Institute, Brazil

LLM PersonalizationSmall Language ModelsCatastrophic Forgetting
AskPDF
Jun 25, 2026

Reproducibility Study of "AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models"

Fang et al. (2025) introduced a null-space constrained projection, named AlphaEdit, for locate-then-edit knowledge editing methods, theoretically guaranteeing that edits do not disrupt previously preserved knowledge, and reports substantial gains over existing editing methods on LLaMA3, GPT2-XL, and GPT-J. In this work, we present a reproducibility study of AlphaEdit, reproducing its reported results under the original experimental setup and extending the evaluation along three axes: new model architectures, additional downstream benchmarks, and substantially longer sequential editing horizons. We successfully reproduce AlphaEdit's reported metrics across the original models, though we identify a discrepancy in the reported fluency and consistency metric. Extending AlphaEdit to newer model families, we find that its advantage does not generalize uniformly, which we trace to architectural assumptions in the locate-then-edit paradigm that are violated by these newer models. We further stress-test AlphaEdit's central sequential-editing claim by extending the number of edits well beyond those evaluated in the original paper, and find that performance, which is stable at the originally reported scale, degrades as edits reach a much higher count, indicating that the null-space projection's protection against catastrophic forgetting is bounded rather than unconditional. Finally, we extend evaluation of edited models on three extra benchmarks, namely, BoolQ, HellaSwag, and XSTest, and we find that large-scale sequential editing degrades both general downstream task competence and safety-relevant refusal behavior. Our results confirm that AlphaEdit performs as reported within its original scope, while showing that its core theoretical guarantees are sensitive to model architecture and editing scale in ways that have practical implications for its deployment.

Ananth K Suresh, Arya Hariharan

Independent

Knowledge EditingModel EditingLanguage Model EditingCatastrophic Forgetting
AskPDF
Jun 25, 2026

From Weights to Features: SAE-Guided Activation Regularization for LLM Continual Learning

Weight-space regularization methods such as Elastic Weight Consolidation (EWC) are the standard approach to catastrophic forgetting in continual learning. However, those methods tend to underperform when applied to large language models. We argue that such underperformance can be partly explained by the ``polysemantic'' nature of large language models: per-weight importance estimates utilized by EWC-style regularization are too coarse and cannot isolate the knowledge that needs protection. In this paper, we propose regularizing instead in the model's activation space, using pretrained Sparse Autoencoders (SAEs) as a monosemantic feature dictionary. From the perspective of constrained optimization, we derive a new loss function that uses the SAE feature dictionary to explicitly balance stability and plasticity, and show that EWC is a special case in the one-sided weight-space penalty setting. Unlike replay-based methods that store or revisit examples from earlier tasks, our method requires no previous-task data after mask construction: current-task data is used to compute a compact SAE feature mask, and only this mask is retained for later training. Further, since the feature space has significantly lower dimensionality than the parameter space, the proposed method is more memory efficient. On the TRACE and MedCL continual learning benchmarks, the method achieves the strongest result among approaches without introducing task-specific architectural components, also surpassing traditional weight-space regularization methods like EWC. Beyond performance comparisons, we provide empirical evidence for the polysemanticity thesis: task-relevant representations are linearly separable in the SAE feature basis but indistinguishable from chance in the weight basis, and weight-space protection is nearly non-selective at the concept level.

Evan Ning, Wei Xue, Dong Lou +1

1The Hong Kong University of Science and Technology

Continual Learning for LLMsCatastrophic ForgettingRegularization-Based Continual Learning
AskPDF
Jun 23, 2026

The Gentle Collapse: Distributional Metrics for Continual Learning

Accuracy degradation is the standard metric for Catastrophic Forgetting (CF), however, it records only whether forgetting occurred or not. It saturates at the extremes and collapses discretely at task boundaries, hiding the internal structure of what is being forgotten. We introduce six softmax-derived metrics spanning true-label rank (TLR), predictive confidence, and distributional divergence that characterize forgetting continuously, each normalized to [0, 1] with no modification to training. On CIFAR-100, these metrics carry information where accuracy does not: at 0% accuracy, the Confusion Margin spans an IQR of [0.32, 0.50] across classes that accuracy treats identically. We demonstrate that this richer signal is actionable in mitigating catastrophic forgetting. Per-sample metric scores used as loss weights reduce forgetting by 1.3 percentage points over uniform experience replay (ER) on CIFAR-100. Furthermore, the slope of a metric over a small window provides a stable sampling criterion: at a small-window size (e.g. 3 epochs), accuracy-trend degrades to 34.79% (std. = 2.32) while log-TLR achieves 41.07% (std. = 0.57). This gap is structural since reliable small-window trend estimation requires a continuous signal. On TinyImageNet, log-TLR trend sampling reduces forgetting by 7.7 percentage points over the ER baseline.

Ahmed Anwar, Andreas Wagner, Federico Raue +2

1German Research Center for Artificial Intelligence (DFKI) · 2RPTU University Kaiserslautern-Landau, Kaiserslautern, Germany · 3Hochschule Karlsruhe University of Applied Sciences, Karlsruhe, Germany

Continual LearningCatastrophic Forgetting
AskPDF