Today for AI
HOT RADAR
llmHEAT 17.0°

AI CURATED HOT STORY · 11/2/2026

Meta's TextSeal2 Boosts AI Watermark Detection to 89% Under Speculative Decoding

1 reports archived1 independent sourcesupdated 11/2/2026, 23:20:56
Synthesis & Latest Updates
1 Sources Cross-Validated

Meta introduces TextSeal2, a new text watermarking algorithm that blends two secret keys into one virtual key with added noise to preserve signal integrity during speculative decoding. Experiments on Qwen3.5-9B show it achieves an 89% detection rate for 200-token outputs at a 0.1% false-positive rate, outperforming previous methods while maintaining generation diversity.

LATEST/Meta introduces TextSeal2, a new text watermarking algorithm that blends two secret keys into one virtual key with added noise to preserve signal integrity during speculative decoding. Experiments on Qwen3.5-9B show it achieves an 89% detection rate for 200-token outputs at a 0.1% false-positive rate, outperforming previous methods while maintaining generation diversity.

Hourly heat curve

3 fully observed hours
Now
17.0
Peak
18.011-03
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first
  1. Rohan Paul (@rohanpaul_ai)T2·78 pts

    Meta's TextSeal2 Boosts AI Watermark Detection to 89% Under Speculative Decoding

    • TextSeal2 uses a dual-key blending strategy to ensure every token carries full watermark signal, avoiding loss from random selection averaging.
    • Optimized for speculative decoding by selecting replacement tokens from draft leftovers, maintaining acceptance rates while embedding watermarks efficiently.
    • On Qwen3.5-9B benchmarks, it improves detection rates by ~18% over TextSeal and ~22% over textGrain.