NeurIPS 2025PastLarge language models
AI That Keeps Up: NeurIPS 2025 Workshop on Continual and Compatible Foundation Model Updates
CCFM
- Submission deadline
- Sep 3, 2025, 16:00 UTCimported from OpenReview — check the website for extensions
- Submission portal
- OpenReview
- Notes
- Topics were auto-suggested and may be imprecise — edits welcome.
Accepted papers (34)
Fetched from OpenReview (v2) on 2026-06-10.
Balancing Synthetic Data and Replay for Enhancing Task-Specific Capabilities
Continual Learning of Domain Knowledge from Human Feedback in Text-to-SQL
Continual Pre-training of MoEs: How robust is your router?
Continuous Self-Improvement of Large Language Models by Test-time Training with Verifier-Driven Sample Selection
CurLL: Curriculum Learning of Language Models
Curriculum Learning as Transport: Training Along Wasserstein Geodesics
Do Language Models Robustly Acquire New Knowledge?
ELLA: Efficient Lifelong Learning for Adapters in Large Language Models
Embedding‑to‑Prefix: Continual Personalization with Large Language Models
EWC-Guided Diffusion Replay for Exemplar-Free Continual Learning in Medical Imaging
Exploring Continual Distillation of Teachers from Different Domains
Exploring The Effectiveness of Test Time Learning In LLMs for Long Contexts
Harnessing Quantum Principles for Parameter-Efficient Continual Learning
HyperAdapt: Simple High-Rank Adaptation
Information-Geometric Perspectives on Merging Variational Foundation Models
IPA: An Information-Preserving Input Projection Framework for Model Adaptation
Mapping Post-Training Forgetting in Language Models at Scale
Per-Axis Weight Deltas for Frequent Model Updates
Pre-training Limited Memory Language Models with Internal and External Knowledge
Probe-Rewrite-Evaluate: A Workflow for Reliable Benchmarks and Quantifying Evaluation Awareness
PTPP-Aware Adaptation Scaling Laws: Predicting Domain-Adaptation Performance at Unseen Pre-Training Budgets
Retrieval Capabilities of Large Language Models Scale with Pretraining FLOPs
Revisiting Warm-Start Training: No Generalization Loss under Standard Training Schemes
RL's Razor: Why On-Policy Reinforcement Learning Forgets Less
Robust LLM Unlearning with MUDMAN: Meta-Unlearning with Disruption Masking And Normalization
Sample-Efficient Parametric Learning from Natural Language
Sculpting [CLS] Features for Foundation Model-Based Class-Incremental Learning
Slim Adaptation Modules: A Simple yet Strong Baseline for Continual Foundation Models
Specialization after Generalization: Towards Understanding Test-Time Training in Foundation Models
TEMPiRL: Foundational Compounding Temporal Drift Theory for Temporal-Graph Adaptation in Large Language Models
Unlearning That Lasts: Utility-Preserving, Robust, and almost Irreversible Forgetting in LLMs
Vocabulary Customization for Efficient Domain‑Specific LLM Deployment
When Data Falls Short: Grokking Below the Critical Threshold
When Less is More: 8-bit Quantization Improves Continual Learning in Large Language Models