**Continual Learning Mechanisms Compose for Long-Horizon Memorization**

> Language models may need to internalize information that arrives over time and retain it through many subsequent updates. To study this challenge, we introduce long-horizon memorization, a setting in which a model learns 100 query-answer tasks through continual supervised fine-tuning without retaining earlier training examples or receiving task identifiers at inference. Sequential updates cause catastrophic forgettin...

**来源信息**
- **来源**：HuggingFace Daily Papers（社区热门论文）
- **分类**：ai-models
- **发布时间**：2026-09-16 08:00（北京时间）
- **原文**：[打开原文](https://huggingface.co/papers/2609.06986)