Replay on Demand: An Emergent Curriculum for Balancing Adaptation and Forgetting in Continued Pretraining
cs.LG
Submitted: 2026-09-30
Updated: 2026-09-30
Code: https://github.com/bethgelab/replay-on-demand
Terminology
Sources
- A Unifying Bayesian View of Continual Learning
- Simple and Scalable Strategies to Continually Pre-train Large Language Models
- NVIDIA Nemotron Nano 2: An Accurate and Efficient Hybrid Mamba-Transformer Reasoning Model
- Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning
- Reuse, Don't Retrain: A Recipe for Continued Pretraining of Language Models
- A Practitioner's Guide to Continual Multimodal Pretraining
- Progress & Compress: A scalable framework for continual learning
- A Sovereign, Open-Source Foundation Model for German and English
- CapTrack: Multifaceted Evaluation of Forgetting in LLM Post-Training
- Functional Regularisation for Continual Learning with Gaussian Processes
- Investigating Continual Pretraining in Large Language Models: Insights and Implications
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks