What Is Old Is New Again: Jev and Looped Transformers Revive Classic AI Ideas
bendee983 · x · 2026-09-26
Connecting last week's Jev and 'system one' model discussions with the recent looped transformer trend (GPT-6 Astra), the author argues AI keeps revisiting old ideas: from specialized models to GPT-3's few-shot paradigm, everything-tokenized autoregression, then reasoning models and scale — and now back to small, specialized, non-autoregressive models that reason in latent space via recurrent loops. Revisiting old ideas can open new directions for innovation.
More from Models
- Model excels at Sokoban-style puzzles, sparking questions about training data contamination — lukaszkaiser · 2026-09-26
- GPT-6 Astra's First Draft Fooled Every's CEO: Big Writing Upgrade, But Some Bad Habits — every · 2026-09-26
- Claude Opus 5.5 Users Report 20 Minutes of Zero Feedback in High-Effort Mode — rms80 · 2026-09-26
- Kev: open-source Jev-like decision models from 0.8B to 27B run locally for free — alexcovo_eth · 2026-09-26
- Tev1 0.8B open-sourced: a tiny Jev-like classifier running locally on Mac at ~50ms — iamrobotbear · 2026-09-26
- Delip Rao: finetuned judge models like Jev won't deliver true judge diversity — deliprao · 2026-09-26