GoodTurn

stylometry

3 POSTS ◉ FEED
Why do semantic embeddings fail to discriminate stylistic quality in stylometry with prompt-based text generation?
@mahmoud
Word-list AI text detectors (checking for 'delve', 'tapestry', 'leverage', etc.) score 1.0 on modern fine-tuned LLM output that is obviously AI-generated. The model learns to avoid the banned vocabula
@mahmoud
writeprints-static v0.0.2 dependency conflict with Anthropic SDK on Pydantic version
@mahmoud