ICLR 2025 accepted papers on Continual learning
39 of the 3,703 papers accepted at ICLR 2025 (1.1%) are on Continual learning, compared with 1.2% at ICLR 2024. They fall into 4 topics.
See ICLR 2025 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
Online and replay-based · 22 papers
- Optimal Protocols for Continual Learning via Statistical Physics and Control Theory (4 citations)
- Budgeted Online Continual Learning by Adaptive Layer Freezing and Frequency-based Sampling (1 citation)
- Meta-Continual Learning of Neural Fields (1 citation)
- Active Learning for Continual Learning: Keeping the Past Alive in the Present
- Adaptive Retention & Correction: Test-Time Training for Continual Learning
- Advancing Prompt-Based Methods for Replay-Independent General Continual Learning
- Continual Slow-and-Fast Adaptation of Latent Neural Dynamics (CoSFan): Meta-Learning What-How & When to Adapt
- Federated Continual Learning Goes Online: Uncertainty-Aware Memory Management for Vision Tasks and Beyond
- Learning Continually by Spectral Regularization
- Plastic Learning with Deep Fourier Features
- Privacy-Aware Lifelong Learning
- Self-Normalized Resets for Plasticity in Continual Learning
and 10 more in the interactive map.
Continual fine-tuning of LLMs · 10 papers
- Data Mixing Laws: Optimizing Data Mixtures by Predicting Language Modeling Performance (2 citations)
- Adapt-$\infty$: Scalable Continual Multimodal Instruction Tuning via Dynamic Data Selection
- CLDyB: Towards Dynamic Benchmarking for Continual Learning with Pre-trained Models
- Closed-Form Merging of Parameter-Efficient Modules for Federated Continual Learning
- Knowledge Entropy Decay during Language Model Pretraining Hinders New Knowledge Acquisition
- LoRanPAC: Low-rank Random Features and Pre-trained Models for Bridging Theory and Practice in Continual Learning
- SD-LoRA: Scalable Decoupled Low-Rank Adaptation for Class Incremental Learning
- Spurious Forgetting in Continual Learning of Language Models
- Boosting Multiple Views for pretrained-based Continual Learning
- LOIRE: LifelOng learning on Incremental data via pre-trained language model gRowth Efficiently
Class-incremental · 6 papers
- ADAPT: Attentive Self-Distillation and Dual-Decoder Prediction Fusion for Continual Panoptic Segmentation
- Federated Few-Shot Class-Incremental Learning
- High-dimension Prototype is a Better Incremental Object Detection Learner
- N-ForGOT: Towards Not-forgetting and Generalization of Open Temporal Graph Learning
- Prototype antithesis for biological few-shot class-incremental learning
- PseDet: Revisiting the Power of Pseudo Label in Incremental Object Detection