Viewpoint: Recursion is More Powerful than Attention, RLMs Express Compaction Easily
lateinteraction · x · 2026-08-16
A technical viewpoint discusses the relationship between recursion and attention mechanisms, arguing that recursion is more powerful than attention in reality. The analogy suggests that compaction is agentic recurrence (RNNs), while recursion (RLMs) resembles agentic attention. Recurrence maintains a constant-size state by compressing the past, whereas attention keeps the full context and decides how to re-process it. RLMs can easily express compaction, but the reverse simulation requires heavy assumptions.
More from Research
- Opus 5 test reveals models understand objects deeply but lack novelty — scaling01 · 2026-08-16
- 18-Model AI Research Experiment: Fable 5 Closes 82% of Human Gap — eliebakouch · 2026-08-16
- Pose Resolution Architecture: Exploring Drive and Curiosity in Online Learning Systems — Calmera · 2026-08-16
- Microsoft study: Agent skills can be harmful, 307 failures reveal pitfalls — coallaoh · 2026-08-16
- Meta Paper: Scaling Laws Hold in Tiny Models — ChengleiSi · 2026-08-16
- Adobe et al. introduce SpyRL, solving subjective RL rewards via a spy game — burkov · 2026-08-16