Continual Learning for LLMs

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

21 papers in the last four weeks, up 250% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 153

Jun 22, 2026cs.LG

Fast and Slow Variational Continual Learning

Continual learning remains a major challenge for modern deep networks, partly because commonly used optimizers lack inherent mechanisms for continual adaptation. One such natural mechanism is fast and slow adaptation to balance stability and plasticity. This mechanism has deep roots in neuroscience and biology, but there is no consensus on how to best incorporate it in commonly used optimizers. Here, we show that this can be easily done via the VCL framework, where past posteriors are used as priors in the future. Our key idea is to incorporate slow adaptation via merging of past posteriors to slow down the drift in the knowledge as learning progresses. The merged posterior is then used as the prior in the VCL update to implement the fast-weight updates. These steps can be seamlessly implemented in the IVON optimizer, whose form and costs are nearly identical to that of Adam. We call this new optimizer the Continual IVON (CoVON) optimizer and show that it not only consistently improves over existing VCL optimizers, but also performs better than other weight-regularization strategies across domain-incremental learning, continual pre-training, and fine-tuning of large language models.

Subarnaduti Paul, Yohan Jung, Mohammad Emtiyaz Khan +3

Faculty of Mathematics and Computer Science, University of Bremen, Bremen, Germany · Department of Computer Science & AI, Jeonbuk National University, Republic of Korea · Department of Computer Science, TU Darmstadt, Darmstadt, Germany +3

Continual Learning for LLMs Deep Learning Optimization Continual Learning Domain-Incremental Learning

Jun 20, 2026cs.LG

SamatNext v0.2-B: An Exploratory Study of RMS-Normalized Hybrid Decoders for Curriculum Retention in Small Code Models

Standard autoregressive Transformer decoders can often exhibit substantial forgetting under sequential fine-tuning on shifting curriculum distributions. This technical report evaluates SamatNext v0.2-B, an experimental 356M-parameter hybrid sequence decoder that alternates Differential-Attention-style layers with DeltaNet-inspired simplified linear-state mixer layers using RMS normalization and output scale calibration. We study the model under a controlled staged Python code curriculum and compare it with a parameter-matched Transformer baseline. In this setting, SamatNext v0.2-B achieves a 100.0% pass rate on the controlled Stage 5 holdout while retaining 98.8% of adjacent Stage 3 semantic behavior and reaching 12.0% on the Stage 2E early syntax holdout. The strongest Transformer baseline reaches 97.6% on Stage 5 but retains only 6.0% of Stage 3 behavior. Both architectures remain weak on long-horizon early-stage retention, so the result should be interpreted as evidence of an altered retention/plasticity tradeoff in this controlled setting, not as a general solution to catastrophic forgetting. Code, model specifications, evaluation scripts, and result tables are provided for independent verification.

Samat Zharassov

Continual Learning for LLMs Decoder-Only Language Models Code Generation Catastrophic Forgetting

Jun 18, 2026cs.CL

Beyond Uniform Forgetting: A Study of Sequential Direct Preference Optimization Across Preference Settings

Aligning language models with human preferences often requires optimising multiple behavioural objectives. A practical approach is to apply these objectives sequentially using preference optimisation methods such as Direct Preference Optimisation (DPO), but it remains unclear whether later training uniformly degrades preferences learned earlier or whether the effect depends on the relationship between objectives. We study sequential DPO across four preference settings covering distributional conflict, multi-attribute interaction, strong safety signal, and compatible response-quality objectives. Using Llama-3.1-8B-Instruct with LoRA adapters, we evaluate all objectives after every stage with a fixed base-model reference. We find that sequential DPO does not produce a single forgetting pattern; preference change ranges from partial degradation to stability, pair-level redistribution, or positive transfer depending on objective relationship, signal strength, and training order. Pair-level analysis using length-normalised policy margins shows that aggregate metrics can mask heterogeneous changes across preference pairs, whereas quartile decomposition reveals that high-confidence pairs can either degrade or improve depending on the setting. Mechanistic diagnostics show that Stage~2 gradients and adapter updates are near-orthogonal to the previous objective across all settings, providing little evidence that direct gradient opposition is the primary driver. These findings suggest that future sequential alignment pipelines should account for objective compatibility and signal strength, rather than assuming that later objectives affect earlier preferences uniformly.

Pranav Bhandari, Nicolas Fay, Amitava Datta +2

1Network Analysis and Social Influence Modelling (NASIM) Lab · School of Physics Maths and Computing, The University of Western Australia · School of Psychological Science, The University of Western Australia +1

Continual Learning for LLMs Direct Preference Optimization Preference Optimization Multi-Objective Language Model Alignment Catastrophic Forgetting

Jun 18, 2026cs.LG

LOKI: Memory-Free Null-Space Constrained Lifelong Knowledge Editing

Lifelong knowledge editing aims to efficiently and sequentially update language models over time, as new knowledge becomes available or when the model makes mistakes, while preserving acceptable performance on past knowledge. One unresolved challenge is that existing methods modify a fixed set of layers for all new knowledge samples, reducing flexibility and increasing catastrophic forgetting. Another is requiring access to previous knowledge and extensive pre-processing to obtain data statistics. To address these challenges, we introduce LOKI, a novel approach that uses dynamic layer selection based on the Hilbert-Schmidt Independence Criterion and projects gradient updates onto the null-space of the model weights, bypassing the requirement for previous knowledge access. We show that LOKI achieves superior performance to existing approaches across a wide variety of experiments, achieving up to a 14% improvement in average accuracy.

Masih Eskandar, Miquel Sirera Perelló, Stratis Ioannidis +1

Department of Electrical and Computer Engineering Northeastern University Boston, MA, 02115

Knowledge Editing Continual Learning for LLMs Model Editing Language Model Editing Catastrophic Forgetting

Jun 17, 2026cs.AI

Repeated post-training is not Self-improving: Diagnosing Scientific Amnesia in Continual DPO Pipelines

Industrial LLM teams often ship behavior updates by repeatedly DPO-training a base model on sequences of related preference-data campaigns. The dominant failure mode in this regime is not always classical catastrophic forgetting: a pipeline may preserve previously learned behaviors while still failing to accumulate reusable methodological knowledge about how to train the next campaign. We call this failure mode scientific amnesia. This paper turns that practitioner intuition into a measurable industrial problem. We contribute: (i) a diagnostic suite for amnesia, (ii) a Program-based pipeline that chains FSDP-sharded DPO checkpoints across Qwen2.5-7B-Instruct runs, (iii) a 30-campaign HumanEval subdomain benchmark, and (iv) a comparative diagnostic study of five strategy proposers: random memory, rule-based scheduling, retrieval-only memory, warm-start Bayesian optimization, and MSCL, a meta-scientific memory and reasoner candidate. Across a single-seed 5-condition * 3-step real-LM chain, 4 of 5 candidates degrade in step-level peak pass@1, including MSCL; only the deliberately conservative rule-based schedule improves. Follow-up pilots qualify rather than overturn this finding: in a heterogeneous chain, MSCL is the only completed candidate that improves, whereas in a small multi-seed homogeneous sweep, retrieval-only has the best mean Delta and no pairwise candidate gap is statistically distinguishable. The contribution is therefore diagnostic, not a claim that MSCL solves the problem: scientific amnesia is observable in a production-like continual-DPO pipeline, and conclusions about interventions depend sharply on chain regime, evaluator design, and seed coverage.

Jianzhe Lin, Fei Wang, Xiaolin Li +2

MetaAI

Continual Learning for LLMs Direct Preference Optimization LLM Post-Training Generative AI Evaluation

Jun 16, 2026cs.LG

Continual Self-Improvement with Lightweight Experiential Latent Memories

Large language models achieve strong reasoning performance by scaling inference-time compute, yet remain fundamentally stateless, discarding the rich, self-produced reasoning traces generated during this process. We investigate whether models can instead learn online from this experience, converting transient computation (reasoning traces) into persistent reusable knowledge, and without external supervision or access to future data. We show that In-Context Learning (ICL) over raw reasoning traces fails to generalize, reflecting a fundamental limitation of token-level reuse: individual traces lack the abstraction needed for transfer, even after refinement (e.g. self-reflection). In contrast, drawing inspiration from recent works on unsupervised reinforcement learning, we find that lightweight per-instance training with self-generated test-time signals (majority voting) as rewards yields substantial gains, often surpassing full-dataset offline training, motivating a shift from raw traces to learned latent representations. Building on this insight, we propose an online method that distills inference-time compute spent on encountered problems into compact modular latent memories capturing the underlying reasoning structure. These memories are stored and retrieved for future inputs, enabling continual improvement while avoiding catastrophic forgetting through modular design. Importantly, our method is highly efficient, parametrized as extremely lightweight soft prompt memories (~0.001% of model parameters) and trained with only a few gradient steps, yet achieving performance competitive with full parametric updates and offline training. Across challenging mathematical reasoning benchmarks, our approach significantly outperforms zero-shot and raw data ICL baselines, while transferring effectively across datasets.

Vaggelis Dorovatas, Nancy Kalaj, Rahaf Aljundi

Toyota Motor Europe · University of Trento

Continual Learning for LLMs Memory-Augmented Language Models RL for Language Model Reasoning Persistent Memory for Language Models Test-Time Training Online Continual Learning+2

Jun 14, 2026cs.CL

DYNA : Dynamic Episodic Memory Networks for Augmenting Large Language Models with Temporal Knowledge Graphs in Continuous Learning

Large Language Models (LLMs) struggle to incorporate new knowledge without forgetting or costly retraining. We propose DYNA, a lightweight framework that augments a frozen LLM with a temporal knowledge graph where events are nodes and temporal relations are directed, timestamped edges. The graph serves as an external, updatable memory. At query time, DYNA retrieves relevant nodes via random walks and centrality measures, then augments the LLM's response. Evaluated on three temporal recall tasks, DYNA reduces catastrophic forgetting by ~7% compared to fine-tuning and improves temporal ordering by ~5% over standard RAG. Higher graph clustering coefficients correlate with better retrieval, showing that graph structure matters. Contributions: (1) episodic memory as temporal KG, (2) retraining-free LLM augmentation, (3) graph properties as predictors of retrieval performance.

Ali Sarabadani, Mahtab Tajvidiyan

Department of Computer Engineering and Information Technology, University of Qom, Qom, Iran

Continual Learning for LLMs Memory-Augmented Language Models Graph Retrieval-Augmented Generation Temporal Reasoning in Language Models Persistent Memory for Language Models Temporal IR+2

Jun 14, 2026cs.CL

Retrievable Gradients: Continual Post-Training Without Cumulative Weight Drift

Continual post-training enables models to absorb emerging knowledge after deployment, but repeatedly updating shared parameters can accumulate weight drift, potentially causing catastrophic forgetting and degrading general capabilities. Retrieval-augmented generation avoids such parameter drift, yet often lacks the depth of parametric knowledge integration. In this paper, we propose ReGrad (Retrievable Gradients), a new paradigm that treats gradients as retrievable units of knowledge. ReGrad pre-computes document-specific gradients offline, stores them in an indexed Gradient Bank, and retrieves only query-relevant gradients at inference time for temporary weight adaptation. However, raw language-modeling gradients are optimized for token-level document reconstruction rather than for query-driven knowledge use. We therefore introduce a bi-level meta-learning objective that reshapes document-derived gradients into generalizable adaptation signals for downstream tasks. Experiments across general and domain-specific settings show that \textsc{ReGrad} outperforms CPT and RAG baselines, enabling scalable and reversible parametric knowledge injection without accumulating weight drift.

Weihang Su, Jiacheng Kang, Jingyan Xu +7

Department of Computer Science and Technology, Tsinghua University

Retrieval-Augmented Generation Continual Learning for LLMs Memory-Augmented Language Models Knowledge Augmentation for Language Models

Jun 12, 2026cs.LG

LLM Evolution as an Industry-Scale Ecosystem: A Lifecycle Perspective on Continual Learning

Continual learning capability is critical for Industrial LLMs, as deployed models must be continuously updated to meet evolving requirements and environments, rather than repeatedly retrained from scratch. However, most existing research focuses on improvements on static benchmarks, failing to capture real industrial needs. In this survey, we reformulate Industrial Continual Learning (ICL) for LLMs as a closed-loop update-and-release problem in a versioned ecosystem, where updates propagate hierarchically to industrial, application-specific models and LLM-powered applications, with capability inheritance and transfer across versions and model families. From this ecosystem perspective, we identify three core challenges: repeated adaptation erodes model plasticity, foundation-model upgrades break capability inheritance, and long-term sustainability is constrained by deployment requirements. We then organize the technical landscape of ICL around five lifecycle design principles: preserving plasticity headroom, treating upgrades as capability transfer, enabling trustworthy continual reinforcement learning, making training recipes self-optimizing, and building accountability as a base layer for long-term iteration. For each principle, we synthesize representative technical directions. Finally, we evaluate the maturity of each principle and its technical components via an evidence-based lens, identify key gaps hindering real-world deployment, and outline a practical ICL deployment blueprint and a pathway for feeding industrial realities back into academic research.

Hao Jiang, Enneng Yang, Guojie Zhu +7

Huawei Technologies Co., Ltd., China · Shenzhen Campus of Sun Yat-sen University, China · Renmin University of China, China

Continual Learning for LLMs Continual Learning

Jun 9, 2026cs.CL

Continual LLM Upcycling: A Predictor-Gated Bank-Wise Sparsity Training Recipe for Dense-to-Sparse LLMs

We study dense-to-sparse continual training as a way to construct channel-sparse large language models from dense checkpoints. Starting from a Qwen2.5-8B dense backbone, we continue training at 32K context and introduce a predictor-gated sparse SwiGLU FFN in the 32K stage. For each token and layer, we use a low-rank predictor to produce FFN-channel routing logits. We then apply a bank-wise top-k rule to retain 16 channels in every 64-channel bank, yielding 4x sparsity in the FFN intermediate activation. Unlike post-hoc sparse inference methods, the routing module is placed on the main language modeling path and optimized during continual training, enabling the dense model to be upcycled into a hardware-oriented sparse model. We report the architecture, training recipe, benchmark performance, and training lessons. We also identify a layer-local long-context failure mode on RULER-CWE and propose a single-layer repair algorithm that substantially improves the affected length range.

Ruixuan Huang, Jinyuan Shi, Hantao Huang +5

1HKUST · 2Moffett AI · 3Independent Contributor

Continual Learning for LLMs LLM Compression Long-Context Language Modeling Structured Sparsity Language Modeling LLM Routing

Jun 7, 2026cs.RO

LUNA-AD: Lightweight Uncertainty-Aware Language Model with Lifelong Learning for Autonomous Driving

While large language models (LLMs) offer promising reasoning capabilities, their integration into safety-critical driving systems is hindered by limited reasoning diversity, high computational overhead, and static learning paradigms. To address these challenges, we propose LUNA-AD, a lightweight uncertainty-aware language model with lifelong learning for autonomous driving (AD). LUNA-AD features a tri-system architecture that reconciles complex multimodal behavioral reasoning, efficient deployment, and continual refinement. We design a multi-agent analytical system to generate uncertainty-aware decision-making demonstrations through diverse hypothesis exploration. A dual-head lightweight heuristic model is distilled to unify the inference of decision distributions and textual explanations while enabling efficient deployment. Furthermore, a reflection-driven lifelong learning mechanism operates on multimodal decision outputs and preserves strategic diversity, allowing for the refinement of candidate decisions and rationales via closed-loop feedback to enhance driving robustness. Extensive experiments on nuPlan benchmarks demonstrate that LUNA-AD achieves state-of-the-art success rates under both non-reactive and reactive modes, with drastically reduced inference latency compared to existing knowledge-driven AD frameworks.

Ruoyu Yao, Pei Liu, Ruiguo Zhong +3

The Hong Kong University of Science and Technology (Guangzhou), China

Continual Learning for LLMs Language Model-Based Control Continual Learning Autonomous Driving Planning Decision-Making under Uncertainty

Jun 5, 2026cs.LG

Sparse Subspace-to-Expert Sharing for Task-Agnostic Continual Learning

Continual learning in Large Language Models (LLMs) is hindered by the plasticity-stability dilemma, where acquiring new capabilities often leads to catastrophic forgetting of previous knowledge. Existing methods typically treat parameters uniformly, failing to distinguish between specific task knowledge and shared capabilities. We introduce Mixture of Sparse Experts for Task Agnostic Continual Learning (SETA), a framework that resolves the plasticity-stability conflict through adaptive sparse subspace decomposition into task-specific expert modules. Unlike standard updates, where tasks compete for the same parameters, SETA separates knowledge into unique experts, designed to isolate task-specific patterns, and shared experts, responsible for capturing common features. This structure is maintained through adaptive elastic anchoring and a routing-aware regularization that jointly protect shared knowledge at both the weight and routing levels and enable a unified gating network to automatically retrieve the correct expert combination during inference. Extensive experiments across diverse domain-specific benchmarks demonstrate that SETA achieves competitive or superior overall performance relative to state-of-the-art continual learning baselines, with particularly strong retention of early-task knowledge and improved backward transfer on LLaMA-2 7B and Qwen3-4B.

Fatema Siddika, Md Anwar Hossen, Tanwi Mallick +1

Iowa State University, Ames, USA · Argonne National Laboratory, USA

Continual Learning for LLMs Mixture-of-Experts Language Models Catastrophic Forgetting

Jun 4, 2026cs.LG

RECAP: Regression Evaluation for Continual Adaptation of Prompts

Production agentic systems routinely face evolving constraints and must comply from the very next interaction. Scenarios like a tool-call notification changing a compliance threshold or a policy update adding disclosure requirements fit this criteria, having close to no room for errors in production. This proactive adaptation setting is common in deployment, but absent from current benchmarks, which assume either static constraint sets or reactive protocols with evaluation feedback. We introduce RECAP, a benchmark that measures continual-learning phenomena (forgetting, regression, forward transfer) at the constraint level under a strictly proactive adapt-then-test protocol: prompt optimization methods receive only the constraint specification and must generalize before seeing any test data. Evaluating six methods across five LLMs and three schedules with evolving constraints, we find that these methods show no significant improvement in performance, even after incurring a higher latency. These methods, designed for offline or reactive settings, are inadequate for the proactive paradigm. Our work emphasizes the growing need for designing proactive prompt adaptation methods, where the models must remain robust to evolving needs in deployment.

Harsh Deshpande, Kushal Chawla, Sangwoo Cho +2

Capital One

Continual Learning for LLMs LLM Evaluation Instruction Following

Jun 4, 2026cs.AI

Continual Learning Bench: Evaluating Frontier AI Systems in Real-World Stateful Environments

Continual learning, the ability of AI systems to improve through sequential experience, has attracted substantial interest, but no high-quality benchmark exists to evaluate it. We introduce Continual Learning Bench (CL-Bench), the first difficult, expert-validated benchmark designed to measure whether LLM-based systems genuinely improve with experience. CL-Bench spans six diverse domains (software engineering, signal processing, disease outbreak forecasting, database querying, strategic game-playing, and demand forecasting), each validated by domain experts and designed so that tasks share a learnable latent structure (codebase layout, disease outbreak dynamics, opponent strategies) that a stateful system can discover online but a stateless one cannot. We evaluate frontier models across several agent architectures, from naive in-context learning (ICL) to dedicated memory systems, introducing a gain metric to isolate learning from prior capabilities. We find that these systems leave headroom for improved continual learning: agents frequently overfit to immediate observations or fail to reuse knowledge across instances, and dedicated memory systems do not fix this -- in fact, naive ICL outperforms systems dedicated to memory management. CL-Bench is the first benchmark to evaluate continual learning across diverse real-world domains with expert-validated tasks and isolate online learning from underlying model capability, showing a need for better continual learning systems.

Parth Asawa, Christopher M. Glaze, Gabriel Orlanski +7

UC Berkeley · Snorkel AI · University of Wisconsin-Madison +1

Continual Learning for LLM Agents Continual Learning for LLMs Continual Learning Online Continual Learning Memory-Augmented Language Model Agents

Jun 3, 2026cs.CL

Rethinking Continual Experience Internalization for Self-Evolving LLM Agents

Experience internalization converts contextual experience from past interactions into reusable parametric capability, offering a promising path toward continual learning in large language models (LLMs). While prior work has predominantly focused on single-iteration transfer, we discover that under multi-iteration experience learning, existing methods suffer from a progressive capability collapse rather than compounding improvement. We systematically examine this failure through three vital dimensions of experience internalization: (1) Experience Granularity: We find that principle-level experience is more durable than instance-level experience, as it effectively abstracts transferable strategies away from trajectory-specific details. (2) Experience Injection Pattern: Our analysis reveals that step-wise injection significantly outperforms global injection by aligning experience with intermediate decision states, a property that is critical for long-horizon tool use. (3) Internalization Regime: We demonstrate that off-policy context-distillation on high-quality teacher trajectories provides a substantially more stable training signal than on-policy context-distillation, which is inherently limited by local corrections on student-induced flawed states. Together, these insights yield a simple yet robust recipe for stable and sustainable experience internalization, providing concrete guidance for engineering self-evolving and continually learning LLMs.

Jingwen Chen, Wenkai Yang, Shengda Fan +7

1Gaoling School of Artificial Intelligence, Renmin University of China · School of Software, Beihang University · 3Meituan

Continual Learning for LLM Agents Continual Learning for LLMs Continual Learning Context Distillation

Jun 2, 2026cs.LG

Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories

The past few decades have witnessed significant advances in the design of machine learning algorithms, from early studies on task-specific shallow models to more general deep Large Language Models (LLMs). Despite showing promising results in tasks that require instant prediction or in-context learning, existing models lack the ability to continually learn and effectively transfer their temporal in-context knowledge to their long-term parameters. Inspired by human learning process, we introduce a ''Sleep'' paradigm that allows the models to continually learn, distill their short-term fragile memories into stable long-term knowledge with replay, and recursively improve themselves with ''Dreaming'' process. In more detail, sleep consists of two stages: (1) Memory Consolidation: an upward distillation process, called Knowledge Seeding, where the memories of a smaller-self are distilled into a larger network to provide more capacity while preserving the knowledge. As a proof of concept, we present a new Generalized Distillation process for {Knowledge Seeding} (i.e., the combination of on-policy distillation with Reinforcement Learning (RL)-based imitation learning); (2) Dreaming: a self-improvement phase, where the model uses RL to generate a curriculum of synthetic data to rehearse new knowledge and refine existing capabilities without human supervision. Our experiments on long-horizon, continual learning, knowledge incorporation, and few-shot generalization tasks support the importance of the sleep stage.

Ali Behrouz, Farnoosh Hashemi, Adel Javanmard +1

Google

Continual Learning for LLMs Persistent Memory for Language Models Replay-Based Continual Learning Language Model Self-Improvement Knowledge Distillation

Jun 2, 2026cs.CL

HyperPatch: Sequential Knowledge Editing Under n-ary Structural Drift

Large Language Models (LLMs) rely on Knowledge Editing (KE) to maintain temporal validity, yet real-world knowledge is inherently n-ary. We demonstrate that in non-stationary environments, sequential updates to complex relations induce N-ary Structural Drift, a phenomenon where the binary reification of n-ary events into triples fractures relational atomicity. This precipitates Structure-Conditioned Knowledge Transfer Failure, a systematic mis-grounding of the retriever frequently misdiagnosed as parametric hallucination. To tackle this, we propose HyperPatch, a parameter-preserving framework that reformulates sequential KE as a stability problem over hypergraph manifolds. HyperPatch preserves event integrity through three phases: (i) Structural Prior Initialization, establishing a topology-aware embedding space via contrastive learning on a Hypergraph Neural Network (HGNN) to capture high-order correlations; (ii) Sequential Topology Editing, utilizing a dual-stage mechanism that employs SimHash-based Topological Alignment for rapid conflict resolution and Topological LoRA Adaptation to track drift without backbone retraining; and (iii) Structure-Conditioned Reasoning, which integrates globally consistent evidence from fused linguistic and structural manifolds. On the MQuAKE-CF and MQuAKE-T benchmarks, HyperPatch achieves relative gains in Hop-wise Accuracy (H-Acc) of 96.24% and 21.06% over the strongest baseline, respectively. Further ablations demonstrate superior reliability under continuous n-ary update streams, whereas the standard KG-based variant suffers H-Acc collapses of up to 88.3% due to structural misalignment.

Yu-Kai Chan, Wen-Sheng Lien, Dong-Ting Yao +4

National Yang Ming Chiao Tung University Hsinchu, Taiwan

Knowledge Editing Continual Learning for LLMs Knowledge Graphs

Jun 2, 2026cs.LG

Learning to Solve, Forgetting to Retain: Correct-Set Turnover in RLVR

Reinforcement learning with verifiable rewards (RLVR) improves the ability of large language model, yet headline accuracy gains often conceal a hidden cost: previously solved problems quietly become unsolvable as training proceeds. We frame this phenomenon as \emph{correct-set turnover}, representing the coupled dynamics of solution acquisition and regression over the mastered set. Under this view, retention becomes an explicit optimization target alongside acquisition. We analytically and empirically establish the \emph{repair-window principle}: the cost of restoring a regressed prompt grows sharply with review delay, defining a low-cost window that standard RLVR pipelines fail to exploit. To address this, we propose \textbf{\method{}}, a retention-aware review mechanism that tracks mastered prompts and periodically reintroduces them to \textbf{remind} the model of previous solutions. By utilizing pre-rollout batch replacement, \method{} incurs zero additional rollout overhead. Evaluated across 20 benchmarks spanning image-text, video, and text-only tasks with Qwen3-VL and Qwen2.5-Math, \method{} consistently improves performance over GRPO, DAPO, and replay baselines, demonstrating robust generalizability across modalities and algorithms.

Chuanyu Qin, Chenxu Yang, Qingyi Si +3

Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China · School of Cyber Security, University of Chinese Academy of Sciences, Beijing, China

Continual Learning for LLMs Reinforcement Learning Replay-Based Continual Learning Reinforcement Learning with Verifiable Rewards RL Post-Training

Jun 1, 2026cs.LG

Repurposing Adversarial Perturbations for Continual Learning: From Defense to Active Alignment

In dynamic environments, large language models need to keep adapting to new tasks, but continual learning often suffers from forgetting, limited transfer, and vulnerability to adversarial perturbations. To address this, we present AdvCL, which repurposes adversarial perturbations as a geometric control signal for stable continual adaptation. AdvCL combines three plug-in modules: Intra-Smooth promotes local smoothness via small adversarial perturbations; Proto-Clip uses similarity clipping to prevent excessive alignment to current task prototype; and Inter-Align applies directional alignment toward previous task prototype to reduce representational gaps. Experiments show consistent gains in both standard performance and robustness, with lower forgetting and stronger transfer. We further analyze key mechanisms by quantifying the sensitivity of Intra-Smooth to perturbation settings and the effect of Inter-Align on task similarity and geometric distance. In summary, the modules provide complementary gains when combined, and each can also be integrated individually into diverse CL paradigms, including replay, regularization, and dynamic architectures, thereby offering a geometric control mechanism for continual learning.

Ran Liu, Min Yu, Mingqi Liu +7

Institute of Information Engineering, Chinese Academy of Sciences · School of Cyber Security, University of Chinese Academy of Sciences · Deakin University +1

Continual Learning for LLMs

Jun 1, 2026cs.CL

Training Prompt Matters: State-Adaptive Optimization for Robust Fine-Tuning

While prompt engineering is instrumental in maximizing the capabilities of Large Language Models (LLMs) during inference, the role of prompts during training remains critically underexplored. Prevailing fine-tuning paradigms typically treat training prompts as mere surface forms, assuming that semantically equivalent instructions yield identical learning outcomes. However, we reveal that this equivalence is deceptive: while paraphrased prompts often lead to comparable in-task performance, they induce drastically different cross-task impacts regarding catastrophic forgetting and generalization. Crucially, these impacts are positively correlated across tasks, indicating the existence of superior prompts that consistently yield better performance. Furthermore, we discover that these superior prompts can be robustly identified by task loss prior to learning. Leveraging these insights, we introduce State-Adaptive Prompt Optimization (SAPO), a lightweight yet effective training strategy that shifts task formulation from a static input to a dynamic, state-adaptive variable. Comprehensive experiments on diverse benchmarks confirm its effectiveness, which significantly mitigates forgetting while improving generalization, achieving substantial performance gains over state-of-the-art methods. These results provide insights into how training prompts shape learning dynamics and offer a practical recipe for robust fine-tuning. Our code is available at https://github.com/Eric8932/SAPO.

Wenhang Shi, Yiren Chen, Shuqing Bian +5

School of Information, Renmin University of China, Beijing, China · Peking University, Beijing, China · Tencent, Beijing, China

Continual Learning for LLMs Prompt Learning LLM Fine-Tuning Automatic Prompt Optimization

May 29, 2026cs.LG

Dynamic Proxy-Mixing: Transferring Replay Controllers from Small to Large Models for Continual Instruction Tuning

Continual instruction tuning updates a language model through a sequence of new domains, yet each update can progressively erode previously learned capabilities and alignment behavior. Replay is the standard mitigation, but fixed replay ratios are inherently limited because the optimal mixture varies with the current domain, the training stage, and the evolving vulnerability of prior behaviors. We propose PROX-YMIX, a framework that learns a dynamic replay controller on a small proxy model and transfers the frozen controller to a larger target. The controller never observes future tasks and constructs its state from normalized validation losses and their temporal dynamics, producing a masked mixture over the current task and accessible replay buffers. Our core empirical hypothesis is forgetting mirroring: task vulnerability rankings remain largely consistent across model scales even when absolute loss magnitudes differ. We validate this assumption empirically before transferring controllers across scales. On LLaMA-3-8B across five continual instruction tuning sequences, PROXYMIX improves average accuracy by 3.4 points, reduces final forgetting by 3.5 points, and raises safety score by 5.8 points over the strongest non-oracle baseline, at roughly 50x lower policy learning cost than Oracle Target RL. The framework is leakage free and architecture independent at the interface level, and we also identify settings where the proxy assumption breaks down, highlighting limitations for robust deployment.

Ibne Farabi Shihab, Fariya Afrin, Anuj Sharma

Department of Computer Science, Iowa State University · Department of Computer Science, Kalinga Institute of Industrial Technology · Department of Civil, Construction & Environmental Engineering, Iowa State University

Continual Learning for LLMs Replay-Based Continual Learning Instruction Tuning

May 29, 2026cs.LG

CRMA: A Spectrally-Bounded Backbone for Modular Continual Fine-Tuning of LLMs

Sequential fine-tuning of large language models forces a choice: let the shared substrate keep learning and accept catastrophic forgetting, or freeze it after task one and foreclose cross-task refinement. Per-task adapter methods (LoRAHub, AdapterFusion, PackNet, Progressive Networks) take the second path. We introduce CRMA (Constrained Residual Mixing Adapter), a residual adapter whose internal mixing matrix M is doubly-stochastic at every forward pass via Sinkhorn normalization, so by Birkhoff's theorem ||M||_2 <= 1 holds by construction -- a structural bound, not a penalty. CRMA's spectrally bounded backbone provides a continuously trained shared substrate that earlier modular methods could not, while preserving their forgetting guarantees. On Mistral-7B across 5 sequential domains and 3 seeds, modular per-task LoRA on a CRMA backbone reduces loss-relative drift from +42.96% +/- 5.5 (naive sequential fine-tuning) to -0.17% +/- 0.17, with disjoint per-seed ranges, and improves prior-task holdout loss by 1.99% +/- 0.54 over a matched frozen-substrate baseline. Three independent experimental setups (Mistral-7B 4-domain controlled ablation, TinyLlama 3-domain contamination-controlled replication, Mistral-7B cross-domain probes at 7B) all show positive backward transfer -- without replay buffers, without growing per-task memory, and without distillation. An inference-time ablation on Gemma-2-9B confirms CRMA mediates access to sequentially trained knowledge: 98/100 vs. 38/100 on the same weights and same questions with only CRMA injection toggled. 867 logged training steps verify ||M||_2 = 1.0 within float32 precision (max deviation 1.2 x 10^-7). The forgetting-prevention effect holds across 1.1B-9.2B parameters and four architecture families.

Kiran Nayudu, Aswini Nutakki, Sai Vinay Naidu +1

ModelBrew AI

Continual Learning for LLMs Fine-Tuning Adapter Tuning Parameter-Efficient Fine-Tuning Catastrophic Forgetting

May 29, 2026cs.CL

Parameter Alignment Mitigates Catastrophic Forgetting in Multilingual Expert Language Models

While continual pretraining~(CPT) is a practical way to extend large language models to new languages, naïve finetuning on targeted data erodes existing capabilities through catastrophic forgetting. Organizing training around language families reduces cross-language interference but cannot alone prevent forgetting of the general knowledge needed for downstream tasks. We link this forgetting to parameter drift in multilingual CPT and present a suite of five layer-aware parameter alignment strategies: hard layer freezing, soft regularization, post-hoc weight reversion, and model merging. We systematically compare our alignment strategies against two unregularized CPT baselines on benchmarks spanning 32 training languages from five language families, plus held-out languages, across four evaluation axes: perplexity, reading comprehension, physical reasoning, and translation. Parameter alignment substantially reduces forgetting at minimal cost to language acquisition: layer freezing and regularization best preserve comprehension, whereas post-hoc reversion yields the strongest translation gains. Together, these results map the acquisition--forgetting frontier for family-expert CPT and offer practical deployment guidelines pairing each strategy to the tasks it best serves.

Sanchit Ahuja, Terra Blevins

Northeastern University

Multilingual Language Models Continual Learning for LLMs LLM Alignment Catastrophic Forgetting

May 29, 2026cs.CL

TRACE: Discovering Task-Specific Parameter via Adaptation-Aware Probing for Continual Fine-Tuning

In real-world deployment, LLMs are often adapted continually across tasks to keep LLMs up-to-date in production, where new fine-tuning should preserve previously learned skills. However, indiscriminately mixing tasks can dilute task specialization, while sequential fine-tuning (full-parameter or low rank adaptation) often causes catastrophic forgetting due to destructive overwriting. Replay-based continual tuning and maintaining separate task-specific adapters can mitigate forgetting, but introduce additional compute, storage, and management overhead. Recognizing the redundancy of LLM parameters for any single task, we reframe continual task adaptation as task-specific parameter discovery via adaptation-aware probing: a short warm-start probe exposes a task's adaptation trace, enabling us to identify and isolate the small subset of parameters essential for each task to mitigate catastrophic forgetting. Building on this view, we introduce TRACE, a novel approach for discovering Task-specific paRameters via Adaptation-aware probing for Continual finE-tuning. We perform a short warm-start fine-tune to derive task-specific core parameters by comparing the warm-started and pre-trained models. Core parameters are identified via two strategies: importance scoring (L

_2

norm and Fisher Information) and specificity analysis (cosine similarity of parameter updates). In continual fine-tuning settings, only the active task's core parameters are updated while others remain frozen, preserving prior knowledge. We conduct extensive experiments across multiple standard benchmarks to demonstrate the superior performance of our proposed method. Additionally, we validate the generalization of our method through a cross-model and scale transferability study, demonstrating a "small-to-large" paradigm that guides the fine-tuning of large-scale models under resource constraints.

Xiaosong Han, Ke Chen, Xindi Dai +7

Key Laboratory of Symbolic Computation and Knowledge Engineering of the Ministry of Education · College of Computer Science and Technology, Jilin University Changchun, China · College of Software, Jilin University Changchun, China +3

Continual Learning for LLMs LLM Fine-Tuning Selective Fine-Tuning Parameter-Efficient Fine-Tuning Catastrophic Forgetting

May 28, 2026cs.LG

The Long-Term Effects of Data Selection in LLM Fine-Tuning

Data selection is increasingly used to reduce the cost of large language model (LLM) fine-tuning, with recent methods prioritizing samples by current utility, diversity, quality, or influence. This paper studies a different question: when fine-tuning occurs over multiple stages, can selection strategies that look optimal now make the model less adaptable later? We introduce a long-horizon view of LLM data selection in which a selector is evaluated not only by immediate task performance, but also by future adaptation speed, forgetting, capability imbalance, and out-of-distribution robustness. We compare representative random, loss-based, gradient-based, diversity-based, quality-based, and utility-diversity selection families under a unified multi-stage protocol. Through controlled experiments designed to instantiate this protocol, we show how short-term selectors can exhibit rank reversal: they improve the current stage while slowing subsequent learning and increasing forgetting. We formalize this behavior as \emph{myopic selection}, provide a simple local analysis of why it can occur, and propose a diagnostic Long-Horizon Aware Selection (LHAS) objective that augments immediate utility with coverage, future-proxy transfer, and anti-concentration terms. The study argues that data selection should be evaluated as a training intervention that shapes the model's learning trajectory, rather than only as a local data-efficiency mechanism.

Yuxin Yang, Aoxiong Zeng, Xiangquan Yang

Shanghai University · East China Normal University

Continual Learning for LLMs LLM Fine-Tuning Data Selection Training Data Selection Selective Fine-Tuning

May 28, 2026cs.LG

Representation Collapse in Sequential Post-Training of Large Language Models

Large language models are now adapted through chains of post-training stages rather than through a single instruction-tuning pass. This paper studies whether such sequential post-training gradually compresses internal representations into low-rank, anisotropic, and homogeneous feature spaces. We define a measurement suite for hidden states, logits, token trajectories, and LoRA updates, and we use it to analyze supervised fine-tuning, preference optimization, safety/refusal tuning, math and code specialization, and long chain-of-thought tuning under controlled stage orderings. The central hypothesis is that excessive representation concentration is not merely a geometric curiosity: it predicts reduced plasticity during later adaptation, weaker out-of-domain generalization, and poorer calibration. We further evaluate lightweight interventions, including mixed-domain replay, feature refresh, representation diversity regularization, and LoRA update decorrelation, as ways to preserve future learnability without giving up the behavioral gains of post-training.

Yichen Liu, Mingyu Chen, Hao Wang +7

Hangzhou Dianzi University · Zhejiang Gongshang University · Ningbo University +1

Continual Learning for LLMs Representation Collapse LLM Post-Training Representation Geometry in Language Models

May 28, 2026cs.LG

Overcoming Forgetting in LLM Fine-Tuning with Evolution Strategies

Evolution Strategies (ES) has recently emerged as a competitive alternative to reinforcement learning (RL) for large language model (LLM) fine-tuning, offering advantages through simplicity, scalability, and inference-only training. However, recent work suggests that ES fine-tuning on new tasks may induce forgetting of prior tasks. First, this paper shows that prior task forgetting (1) is better characterized as performance drift rather than irreversible forgetting, with prior-task performance often recovering during ES training; and (2) is not a specific failure mode of ES, but can also arise for fine-tuning with RL methods. Second, it analyzes when and why such drift arises, highlighting its dependence on ES training dynamics, particularly random walk behavior in weakly constrained directions of the weight space. Third, based on these insights, it introduces Anchored Weight Decay (AWD) as a parameter-space regularization technique that constrains optimization toward the initial model parameters. AWD effectively stabilizes prior-task performance while preserving target-task performance, achieving benefits comparable to large ES population sizes at much lower computational cost. Thus, contrary to previous beliefs, the paper shows that prior-task forgetting under ES is largely avoidable, positioning ES as a promising approach for continual learning in LLMs.

Kajetan Schweighofer, Conor F. Hayes, Roberto Dailey +2

Cognizant AI Lab · UT Austin & Cognizant AI Lab

Continual Learning for LLMs Evolutionary Optimization Evolution Strategies Catastrophic Forgetting

May 28, 2026cs.AI

Make LLM Learn to Synthesize from Streaming Experiences through Feedback

Large language models (LLMs) have been widely adopted for synthetic data generation, significantly reducing annotation costs. However, most existing studies treat synthesis as a set of isolated tasks and overlook a more fundamental question: whether a model can learn to synthesize by accumulating experience from past tasks and transferring it to future ones. In this work, we introduce StreamSynth, a new setting in which synthesis tasks arrive sequentially and experience from historical tasks provides informative signals for future synthesis. To address this setting, we propose SynLearner, a general framework that enables synthesis models to acquire reusable synthesis experience over a task stream. Instead of generating data independently for each task, SynLearner encourages the model to explore diverse synthesis patterns, learn from feedback, and balance sample quality with set-level diversity as tasks evolve. Extensive experiments across multiple benchmarks show that SynLearner effectively leverages experience from earlier tasks to improve synthesis performance on later ones, exhibiting consistent cross-task transferability. These findings provide evidence for the feasibility of StreamSynth and highlight synthetic data generation as an experience-driven process that can benefit from task streams.

Zhenlin Hu, Yan Wang, Zhen Bi +7

Huzhou Normal University · Alibaba Group · Zhejiang Key Laboratory of Intelligent Education Technology and Application +1

Continual Learning for LLMs Synthetic Data Generation Synthetic Data Generation for Language Models

May 28, 2026cs.CL

Leveraging Routing Dynamics in Mixture-of-Experts Models for Efficient Language Adaptation

Mixture-of-Experts (MoE) models are widely used to scale language models, yet their expert routing behavior and adaptation in a multilingual setting remain underexplored. In this work, we study multilingual routing dynamics during continual pre-training of an English-centric MoE model on a multilingual corpus, analyzing how expert usage varies across languages. We find that continual multilingual pre-training leads to diffused, language-agnostic routing in early and middle layers, with language specialization primarily emerging in the final layers. We also show that token-level vocabulary overlap between languages plays an important role in how languages are routed. Motivated by these findings, we propose a parameter-efficient adaptation strategy that updates language-specific and shared experts in the final MoE layers. Experiments on MultiBLiMP and Belebele show that our method achieves a strong performance-efficiency trade-off, attaining competitive performance relative to fine-tuning complete final layers, while updating less than 2% of the parameters. Overall, our findings provide insights into where and how language specialization emerges in MoEs during continual pre-training and provide practical insights for low-resource multilingual adaptation. Our code is available at https://github.com/aditi184/moe-routing-adaptation.

Aditi Khandelwal, Marius Mosbach, Verna Dankers +2

Mila – Quebec AI Institute & McGill University

Multilingual Language Models Continual Learning for LLMs Mixture-of-Experts Language Models LLM Routing Parameter-Efficient Fine-Tuning

May 28, 2026cs.CL

Mask the Target: A Plug-and-Play Regularizer Against LoRA Forgetting

Low-Rank Adaptation (LoRA) has become one of the most widely used fine-tuning mechanisms for adapting large language models to new domains, tasks, and users. Yet adaptation performance alone can obscure an important failure mode: LoRA updates may improve performance on the target distribution while degrading prior capabilities learned during pretraining and alignment. We show that this forgetting becomes especially severe when the adaptation distribution differs substantially from the models original training or alignment distributions. The challenge is amplified in practical settings, where the original training and alignment data are typically unavailable. Motivated by this constraint, we study how LoRA based adaptation balances new learning against forgetting in a replay-free setting, and introduce a simple output space regularizer that can be added directly to existing training pipelines. Our method removes the ground-truth token from both the base and adapted model distributions, renormalizes the remaining probabilities, and applies KL regularization only over the non-target vocabulary. This preserves the base models relative preferences among alternative tokens without directly opposing the cross-entropy signal required for adaptation. As the regularizer acts only at the loss level, it requires no replay data, architectural changes, adapter redesign, or inference-time overhead, and can be applied directly to existing LoRA variants. Across all LoRA variants tested and across various backbones, our method improves the frontier between new learning and forgetting when the adaptation distribution differs substantially from the base models original training or alignment distributions, suggesting a broadly applicable route toward more reliable LLM updating.

Runze Xu, Arpit Garg, Hemanth Saratchandran +1

Australian Institute for Machine Learning Adelaide University Adelaide SA 5000

Continual Learning for LLMs LLM Fine-Tuning Low-Rank Adaptation Catastrophic Forgetting