CVPR 2026 accepted papers on Continual learning
59 of the 4,042 papers accepted at CVPR 2026 (1.5%) are on Continual learning, compared with 1.7% at CVPR 2025. They fall into 4 topics.
See CVPR 2026 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Class-incremental · 23 papers
- Beyond Prompt Degradation: Prototype-guided Dual-pool Prompting for Incremental Object Detection
- From Selection to Scheduling: Federated Geometry-Aware Correction Makes Exemplar Replay Work Better under Continual Dynamic Heterogeneity
- HyCal: A Training-Free Prototype Calibration Method for Cross-Discipline Few-Shot Class-Incremental Learning
- IncreFA: Breaking the Static Wall of Generative Model Attribution
- Parameterized Prompt for Incremental Object Detection
- Quantum-Gated Task-interaction Knowledge Distillation for Pre-trained Model-based Class-Incremental Learning
- TALON: Test-time Adaptive Learning for On-the-Fly Category Discovery
- Temporal Imbalance of Positive and Negative Supervision in Class-Incremental Learning
- Unlocking Positive Transfer in Incrementally Learning Surgical Instruments: A Self-reflection Hierarchical Prompt Framework
- AdaPrior: Bayesian-Inspired Adaptive Prior Correction for Long-Tailed Continual Learning
- An Optimal Transport-driven Approach for Cultivating Latent Space in Online Incremental Learning
- Decouple Your Discovery and Memory in Continual Generalized Category Discovery
and 11 more in the interactive map.
Continual fine-tuning of LLMs · 15 papers
- Affordance-First Decomposition for Continual Learning in Video-Language Understanding
- CGL: Advancing Continual GUI Learning via Reinforcement Fine-Tuning
- DSCA: Dynamic Subspace Concept Alignment for Lifelong VLM Editing
- Forging a Dynamic Memory: Retrieval-Guided Continual Learning for Generalist Medical Foundation Models
- Harmonious Parameter Adaptation in Continual Visual Instruction Tuning for Safety-Aligned MLLMs
- Octopus: History-Free Gradient Orthogonalization for Continual Learning in Multimodal Large Language Models
- On Token's Dilemma: Dynamic MoE with Drift-Aware Token Assignment for Continual Learning of Large Vision Language Models
- Sparse Spectral LoRA: Routed Experts for Medical VLMs
- The Blind Spot of Adaptation: Quantifying and Mitigating Forgetting in Fine-tuned Driving Models
- DGS: Dual Gradient and Semantic-Shift Guided Low-Rank Adaptation for Class Incremental Learning
- PACT: Phase-Like Transition Constraints in Adapter-Based Continual Learning of Vision-Language Models
- pH-Strips for Selective Forgetting: A Blunt but Fast Diagnostic Baseline for Machine Unlearning
and 3 more in the interactive map.
Online and replay-based · 14 papers
- A Faster Path to Continual Learning
- Continual Learning for fMRI-Based Brain Disorder Diagnosis via Functional Connectivity Matrices Generative Replay
- Elastic Weight Consolidation Done Right for Continual Learning
- Exemplar-Free Continual Learning for State Space Models
- Is Parameter Isolation Better for Prompt-Based Continual Learning?
- Lifelong Imitation Learning with Multimodal Latent Replay and Incremental Adjustment
- Multimodal Continual Instruction Tuning with Dynamic Gradient Guidance
- $\phi$-DPO: Fairness Direct Preference Optimization Approach to Continual Learning in Large Multimodal Models
- Beyond Myopic Alignment: Lookahead Optimization for Online Class-Incremental Learning
- Beyond the Static-World: Lifelong Learning for All-in-One Medical Image Restoration
- CHEEM: Continual Learning by Reuse, New, Adapt and Skip - A Hierarchical Exploration-Exploitation Approach
- Parameter-efficient Continual Learning for Enhancing Plasticity without Forgetting under Limited Model Capacity
and 2 more in the interactive map.
Domain and prompt-based · 7 papers
- DK-DDIL: Adaptive Knowledge Retention for Dynamic Domain-Incremental Learning in Medical Imaging
- Prompt-Anchored Vision-Text Distillation for Lifelong Person Re-identification
- Cross-Architecture Adaptation: Cloud-Edge Continual Test-Time Adaptation with Dynamic Sampling and Heterogeneous Distillation
- Dynamic Magic: Unleashing Restricted Knowledge for Lifelong Person Re-Identification
- GeCo-SRT: Geometry-aware Continual Adaptation for Cross-Task Sim-to-Real Transfer
- Hyperbolic Prototype Learning with Uncertainty-Aware Consistency for Continual Test-Time Segmentation
- SAME: Sparse and Anchored Model Editing for Heterogeneous Incremental Learning under Limited Data