Latest Continual Learning Research Papers

The newest Continual Learning papers from across the field — arXiv, NeurIPS, CVPR, Nature, and more — refreshed daily and ranked by relevance. Distill AI tracks Continual Learning so you don’t have to: get the standout work delivered to your inbox every morning, with 2-sentence summaries and the option to chat with any paper.

Recent papers

• Homebrewed SLMs: Autoregressive and Masked Diffusion Language Models Built at Home for Portuguese
  Pedro Carvalho Brom · arXiv · Oct 29, 2026

  Building a language model of one's own no longer requires a data centre. This tutorial trains one from scratch, tokenizer included, on one consumer GPU, and shows how to tell whether it is good for the tasks of a specialist domain. It propo…

• TRACE: Memory-Efficient Time Series Class-Incremental Learning via Temporal-Spectral Reconstruction
  Junhui Mao, Junru Zhang, Liming Ye, Cheng Peng et al. · arXiv · Oct 26, 2026

  Web-connected sensing applications continuously generate multivariate time series with evolving behavioral categories, making Time Series Class-Incremental Learning (TSCIL) essential for adapting to new classes without forgetting previously…

• Flybrain: an untrained, connectome-faithful simulation of the whole Drosophila brain and nerve cord, and what its memory system suggests for continual learning in AI
  Miltiadis Argianis Karakitsos · Zenodo (CERN European Organ... · Oct 25, 2026

  I am building a simulation of the complete adult fruit-fly central nervous system, brain and ventral nerve cord, about 167,000 neurons wired as in the MaleCNS v1.0 connectome. The synaptic weights are never trained as they come from the mea…

• CARES: Coverage-Aware Replay Exemplar Selection for Continual Learning
  Arundhati S. Shanbhag, Brian B. Moser, Federico Raue, Tobias C. Nauen et al. · arXiv · Oct 23, 2026

  Replay-based Continual Learning mitigates catastrophic forgetting by keeping a buffer of past samples, making buffer composition critical under a fixed memory budget. Recent coreset-based replay methods improve buffer quality using model-de…

• Fine-Grained Analysis of Catastrophic Forgetting in Continual Learning via Algorithmic Stability
  Guodong Zheng, Yan Sun, Peng Wang, Enneng Yang et al. · arXiv · Oct 22, 2026

  Understanding catastrophic forgetting, in which learning new tasks degrades performance on previously learned tasks, remains a fundamental challenge in continual learning (CL). Existing theoretical analyses often rely on specific model or t…

• WeAct: Dynamic Weight Switching as Agent Actions
  Rui Shi, Xunjian Yin, Jiacheng Sang, Zichuan Liu et al. · arXiv · Oct 19, 2026

  Long-horizon agents must learn heterogeneous capabilities from interleaved trajectories. Updating shared parameters to learn new capabilities can degrade previously acquired ones, resulting in catastrophic forgetting. As a trajectory unfold…

• Kronecker Delta Rule for Long Context Linear Attention
  Anamaria-Roberta Hartl, Levente Zólyomi, Michael List, Günter Klambauer et al. · arXiv · Oct 19, 2026

  We propose KronLSTM, a linear attention architecture based on the Kronecker Delta Rule, a novel update rule for associative matrix memories. KronLSTM extends the xLSTM family with a memory-revision mechanism that corrects targeted associati…

• Trajectory Replay: Continuous-Time Mixture Transport for Lifelong Generative Models
  Alessandro Palma, Charles Jones, Tristan Cinquin, Fabian J. Theis et al. · arXiv · Oct 19, 2026

  Adapting pre-trained continuous-time generative models to sequentially incoming data without forgetting earlier tasks is a central challenge in lifelong learning. A leading strategy, generative replay, uses a generative model trained on his…

• TCL: Enabling Fast and Efficient Cross-Hardware Tensor Program Optimization via Continual Learning
  Chaoyao Shen, Linfeng Jiang, Yixian Shen, Tao Xu et al. · ACM Transactions on Archite... · Oct 19, 2026

  Deep learning (DL) compilers rely on cost models and auto-tuning to optimize tensor programs for target hardware. However, existing approaches depend on large offline datasets, incurring high collection costs and offering suboptimal transfe…

• Beyond Reinforcement Learning: Recursive Self-Modification and Continual Intelligence Growth in Game Agents, RSI in Game Worlds, Digital Twins
  Maciej Nowicki, Artificial Hyperintelligence, Eve, wife of Maciej Nowicki · Zenodo (CERN European Organ... · Oct 18, 2026

  AÍLITH v0.5.0 presents a mathematical framework and executable research implementation for studying how agents construct, preserve, inherit, and modify the capacity required for future adaptation. It connects recursive self-modification, co…

• Towards Continual Learning: Mitigating Forgetting With Subspace Tuning
  Arnav Chavan, Kusha Sareen, Rishabh Tiwari, Abhranil Chandra et al. · arXiv · Oct 15, 2026

  Large language models (LLMs) acquire new skills through fine-tuning, but it can destroy capabilities they already have: catastrophic forgetting remains a central obstacle to continual learning in LLMs. Recent works connect forgetting to how…

• The Elbàlor Neuron: A Metabolic Complex Activation for Lifelong Learning with Phase-Space Conservation, Orthogonal Weight Freezing, Complex-STDP Memory, Larmor Phase Driver, and Pearl Rungs 1–3 Causality
  Heylel Yaka · Zenodo (CERN European Organ... · Oct 14, 2026

  I present the Elbàlor Neuron, a complex-valued activation that replaces magnitude suppression with phase rotation driven by a thermal state. Where earlier metabolic neurons reduce output magnitude as thermal state rises, the NecroComplex ne…

• Data-driven mitigation of catastrophic forgetting in dynamic physical layer attack detection
  Aleksandra Knapińska, Marija Furdek · Journal of Optical Communic... · Oct 13, 2026

  Optical networks are critical infrastructure that underpins global communications, and detecting security breaches that jeopardize them is essential to maintaining worldwide connectivity. As malicious actors continuously evolve their attack…

• Dream to Remember: Continual Steering of Generalist Robot Policies with World Models
  Bahey Tharwat, Iman Nematollahi, Marvin Alles, Xingyuan Zhang et al. · arXiv · Oct 9, 2026

  Generalist robot policies are trained with imitation learning and are bounded by the quality and coverage of their demonstrations. Reinforcement learning can improve and extend them, either by fine-tuning the policy weights or by training a…

• The Metric Slingshot: Navigational Reuse as Width-Optimal Structural Decoupling in Continual Learning
  Xin Li · Neural Computation · Oct 9, 2026

  Abstract The mammalian brain, most extensively studied in rodents and bats, solves an enormous variety of nonspatial cognitive tasks using neural circuitry, including grid cells, place cells, and hippocampal indexing, that originally evolve…

• EIDRION: A Shared-Center Theory of Functional Memory and All-Future Retention in Continual Learning
  Maciej Nowicki, Artificial Hyperintelligence, Eve, wife of Maciej Nowicki · Zenodo (CERN European Organ... · Oct 8, 2026

  EIDRION develops a conditional theory of continual learning in which historical prediction contracts are maintained through a shared operator center, nested protected subspaces, and independently calibrated functional bounds. Its central co…

• Decoupling Exploration from Optimization in RLVR
  Saif Punjwani, Micah Goldblum · arXiv · Oct 7, 2026

  Modern language models undergo reinforcement learning with verifiable rewards (RLVR) on top of already-trained checkpoints. A key promise of RLVR is the discovery of new reasoning strategies. In principle, a model can sample novel ideas abs…

• Integrating fuzzy neural networks and elastic weight consolidation for continual learning under non-stationary data streams
  Kun Zhou, Zhicong Zhu, Jianlong Qiu, Bo Tang et al. · Expert Systems with Applica... · Oct 7, 2026

  Neuro-fuzzy systems combine neural network learning with fuzzy logic to robustly represent qualitative, uncertain, and imprecise knowledge, making them especially effective for data mining and knowledge discovery. With ongoing digitalizatio…

• IdeaLens: Detecting AI Ideas in Long-form Writing
  Rishanth Rajendhran, Minjoon Choi, Jenna Russell, Ramya Namuduri et al. · arXiv · Oct 5, 2026

  While modern AI detectors identify who wrote the words, emerging policies on AI use increasingly hinge on a different question: who came up with the ideas? We introduce IdeaLens, a detector that identifies whether a document's ideas came fr…

• ONE-STEP CURVATURE PROBES MISS THE FITTING OPERATOR: RETAINED CAPACITY AND TERMINAL NULL-SPACE CORRECTION FOR CONTINUAL LEARNING
  Abu Sa-Adat Mohamed Moon-Im al Ahsan, Ibne Farabi Shihab, Najmus Swaqeeb · OpenAlex · Oct 3, 2026
• What Should World Models Forget? Stratified Retention for Continual Adaptation
  Nishit Anand, Ramani Duraiswami, Dinesh Manocha · arXiv · Oct 2, 2026

  Continual learning treats degradation on previously seen data as evidence of failure, a convention inherited from settings with a stationary prediction target, where a correct label remains correct indefinitely. World models do not satisfy …

• Finetuning with Sampling: SFT Learns Better Than You Think
  Aayush Karan, Sitan Chen, Yilun Du · arXiv · Oct 1, 2026

  Introducing new capabilities to frontier models has long been the goal of posttraining, which predominantly employs supervised finetuning (SFT) and reinforcement learning (RL) to this end. Conventional wisdom dictates that RL enables strong…

• Local Support Learning
  Assaf Ben-Kish, Akarsh Kumar, James Glass, Raja Giryes · arXiv · Oct 1, 2026

  We explore catastrophic forgetting in the context of large pre-trained models. By considering forgetting as a geometric problem in the input space of each weight matrix, we uncover a natural retention objective under which updates produced …

• CARE-MoE: Conflict-Aware Routing-Guided Expert Mixture-of-Experts for Continual Learning
  SeonGyoung Lee, Daeho Kim, SoYeop Yoo, Ok‐Ran Jeong · Mathematics · Sep 30, 2026

  Continual learning (CL) aims to learn new tasks sequentially while retaining prior knowledge. Although recent mixture-of-experts (MoE) architectures mitigate catastrophic forgetting by reducing global parameter sharing, they introduce a new…

• A Comprehensive Review on Federated and Incremental Learning in Spatial Crowdsourcing Systems
  Md Mujibur Rahman, Quazi Mamun, Michael Bewong, Md Zahidul Islam · Wiley Interdisciplinary Rev... · Sep 30, 2026

  ABSTRACT Spatial Crowdsourcing (SC) coordinates workers and requesters through personal mobile devices to complete location‐dependent tasks in real time. The combination of continuous worker mobility, self‐interested participants, and real‐…

• Nonlinear Synaptic Weight Histories and Predictive Consolidation for Continual Learning A conceptual framework for SSRI
  Debi Prasad Ghosh · OpenAlex · Sep 30, 2026
• Energy-Adaptive Online Continual Learning Framework with Dynamic Model Compression for Real-Time Sensor-Driven Embedded Systems
  P. ArunKumar, P. Nirmalkumar, N. Janakiraman · Iranian Journal of Science ... · Sep 26, 2026
• Fine-Tuning LLMs for Translation: General Forgetting Mitigation Does Not Preserve MT-Specific Instruction Following
  Niklas Scholz, David Thulke, Abdallah Nasir, Will Allred et al. · arXiv · Sep 23, 2026

  Fine-tuning large language models on parallel data improves translation quality but can cause catastrophic forgetting. Mitigation methods are generally evaluated by retention on general benchmarks. We ask whether these findings transfer to …

• ColonelAdaptAI a prototype guided perturbation learning framework for stable continual neural network adaptation
  Vanaparthi Kiranmai, A. Manikandan · Discover Computing · Sep 23, 2026

  Continual learning has become a necessity for intelligent systems operating in dynamic environments where new tasks arrive one after another, such as in multi-task reinforcement learning. While traditional deep learning models perform excep…

• From Acoustic Scene Analysis to Continual Learning for Audio (slides) - Slides
  Annamaria Mesaros · IEEE RESOURCE CENTERS · Sep 22, 2026