EMNLP paper: AI-generated text has a consistent stylometric fingerprint across models and domains
najoungkim · x · 2026-09-05
An EMNLP 2026 Main conference paper finds that AI-generated text exhibits a consistent stylometric footprint across different models and domains, while AI-edited text shows a distinct, different pattern. The distinction matters for AI text detection and provenance: generation and editing leave separable stylistic signatures.
More from Research
- Google and HHMI map the complete male fruit fly brain: 166,000 neurons, 125M synapses — ymatias · 2026-09-05
- 2D-RoPE for image models is easier in the complex representation, says arohan — _arohan_ · 2026-09-05
- arXiv paper weighs whether 'AI psychosis' should be a distinct clinical entity — gerardsans · 2026-09-05
- Neural Action Codec: compressing robot actions like audio to train VLA policies — YuXiang_IRVL · 2026-09-05
- Extropic unveils Z1T sparse models claiming up to 140x energy efficiency gains over GPUs — beffjezos · 2026-09-05
- Tencent paper: environment evolution generates harder RL environments without watching the agent — omarsar0 · 2026-09-05