New Study: Training a Single Transformer Layer Can Match Full-Parameter RL
tokenbender · x · 2026-08-16
arXiv paper 'Is One Layer Enough?' finds that training a single transformer layer can recover most of the gains of full-parameter RL training, sometimes surpassing it. Across 7 models, 3 RL algorithms, and multiple task domains, high-contribution layers concentrate in the middle 40-60%.
More from Research
- ImageNet should retire: DiT method rankings negatively correlate between ImageNet and text-to-image — YouJiacheng · 2026-10-16
- GAE paper learns a geometry-native latent space for 3D-consistent world generation — yshan2u · 2026-10-16
- Agents turn conservative as context grows: edits shrink from 150 to 30 lines — dbreunig · 2026-10-16
- LMR-BENCH at EMNLP 2025 tests whether LLM agents can reproduce real LLM research — ZhiyuChen4 · 2026-10-16
- Ex-OpenAI researcher claims most neurons are legible now; critics call it hype — anshulkundaje · 2026-10-16
- Physicist pushes back on ML community's acceptance that models may be beyond understanding — tszzl · 2026-10-16