ICML 2023 accepted papers on Continual learning
21 of the 1,865 papers accepted at ICML 2023 (1.1%) are on Continual learning, compared with 1.1% at ICML 2022. They fall into 3 topics.
See ICML 2023 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 · 13 papers
- Theory on Forgetting and Generalization of Continual Learning (12 citations)
- BiRT: Bio-inspired Replay in Vision Transformers for Continual Learning (5 citations)
- Parameter-Level Soft-Masking for Continual Learning (5 citations)
- Do You Remember? Overcoming Catastrophic Forgetting for Fake Audio Detection (4 citations)
- DualHSIC: HSIC-Bottleneck and Alignment for Continual Learning (4 citations)
- Towards Robust Graph Incremental Learning on Evolving Graphs (2 citations)
- Continual Learning in Linear Classification on Separable Data (1 citation)
- Continual Task Allocation in Meta-Policy Network via Sparse Prompting (1 citation)
- Does Continual Learning Equally Forget All Parameters? (1 citation)
- The Ideal Continual Learner: An Agent That Never Forgets
- Adaptive Compositional Continual Meta-Learning
- Neuro-Symbolic Continual Learning: Knowledge, Reasoning Shortcuts and Concept Rehearsal
and 1 more in the interactive map.
Continual fine-tuning of LLMs · 4 papers
- Lifelong Language Pretraining with Distribution-Specialized Experts (6 citations)
- Discrete Key-Value Bottleneck (4 citations)
- Continual Learners are Incremental Model Generalizers (1 citation)
- Master-ASR: Achieving Multilingual Scalability and Low-Resource Adaptation in ASR with Modular Learning (1 citation)