# BAAI's CURE adds 0.7% compute overhead to diffusion LLM safety

> Original title: The best part? Test-time alignment doesn't need to double your compute budget:

- Source: BAAI
- Published: 2026-10-29T13:26:01.000Z
- Bad Influence: https://badinfluence.news/items/tmfr4hf80fbn3dcmpemewdhk2
- Original: https://x.com/BAAIBeijing/status/2115797302212256148

## Summary

BAAI says its CURE test-time alignment method adds 0.84 TFLOPs, or 0.7%, to LLaDA-8B-Instruct's 126.29 TFLOPs of base generation. Prior test-time defenses SRIGuard and DiffuGuard add 200.0% and 103.1%, which the post puts at 150x to 300x more extra compute. BAAI says it is extending intermediate-state editing toward semantic-level reasoning repair and factuality and logical-consistency control.

## Post

The best part? Test-time alignment doesn't need to double your compute budget:

⚡ 0.7% Extra Compute Overhead:

On LLaDA-8B-Instruct (base generation: 126.29 TFLOPs), prior test-time defenses like SRIGuard and DiffuGuard add +252.58 TFLOPs (+200.0%) and +130.24 TFLOPs (+103.1%)—whereas CURE adds just 0.84 TFLOPs (+0.7%), cutting extra compute by 150x–300x!

🌐 Towards Proactive Generation Control:

Beyond token-level safety, BAAI is extending intermediate-state editability toward semantic-level reasoning repair and multi-constraint test-time control (factuality & logical consistency)—advancing AI safety from passive output filtering to proactive, trustworthy generation control.

📄NeurIPS：https://neurips.cc/virtual/2026/loc/sydney/poster/151876

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHVyyynxbMAA1Dj7.png&mode=thumb&exp=1793491200&sig=47e9e809e083a0b2)