diff --git a/index.html b/index.html
index b5d0f0e..8b23ebd 100644
--- a/index.html
+++ b/index.html
@@ -632,11 +632,14 @@
- 🔥 RLAIF algorithms: PPO, GRPO, SPO (native PyTorch)
+ - Checkpoint resume training: auto-save & cross-GPU recovery
+ - RLAIF dataset: rlaif-mini.jsonl (10K samples); Simplified DPO dataset with Chinese data
- YaRN algorithm for RoPE length extrapolation
- Adaptive Thinking in reasoning models
- Tool Calling & Reasoning tags support
+ - Complete RLAIF chapter with training curves
- SwanLab integration (WandB alternative for China)
- - Bug fixes and performance improvements
+ - Code standardization & bug fixes