TurnTurk-GPT: GPT-2 Fine-Tune Beats Multilingual End-of-Turn Detectors for Turkish
Ahmet Tuğrul Bayrak · hf · 2026-11-02
A paper introducing TurnTurk-GPT, a text-based turn-taking predictor for Turkish built by fine-tuning Turkish GPT-2 with the TurnGPT objective on synthetic and real dialogue.
- Evaluated on real Turkish conversations at word-level decision points on unpunctuated, lowercased streaming-ASR-style text, with recording-disjoint and speaker-disjoint splits, nested threshold selection, and paired cluster bootstrap tests
- Outperforms a widely used multilingual end-of-turn detector under both splits
- Findings show Turkish end-of-turn cues rely on verb-final, agglutinative morphology, differing substantially from English
More from Research
- Astra IDs Breakout, Pong, Pac-Man and Freeway from raw model weights alone — burny_tech · 2026-11-03
- World model that predicts the future directly in 3D, fully open-sourced — YunzhuLiYZ · 2026-11-03
- Scalable gradient attribution finds deceptively harmful training data, with a 0.6B model as monitor — PeterHndrsn · 2026-11-03
- Dynamical releases SDL-2: frontier agents make correct materials-science calls in only ~half of experiments — BenBlaiszik · 2026-11-03
- Follow-up: Astra's weight edits only worked on Qwen's visible flaw, not on Gemma 270m — teortaxesTex · 2026-11-03
- Zuckerberg's Biohub bets $1.8B on virtual cells: the bottleneck is data, not AI — HarperSCarroll · 2026-11-03