Three selection criteria separate at different rates — LOO-CV is inconsistent
BlackHC · x · 2026-09-09
Quantitative result from the thread: with equal loss floors, models separate at three different rates (Δλ is the model complexity delta):
- Final height: gap Δλ/N — LOO-CV is inconsistent (Shao 1993)
- Last-half tail: Δλ·log 2 — constant
- Total area: Δλ·log N — grows without bound
Conclusion: "best model" is incomplete until you specify which part of learning and deployment you value. The full post includes the exact identity, flip-order results, crossing regimes, a toy experiment, and an interactive panel.
More from Research
- EdgeRock open-sourced: 72 MiB KV cache at 256K context, 99.1% smaller than Qwen3.5-4B — alifcoder · 2026-10-27
- EdgeRock-4B holds KV cache at 72 MiB across 256K context, 99.1% smaller than Qwen3.5-4B — alifcoder · 2026-10-27
- TRL adds Score Centering, a simple fix for RLHF training instability — lvwerra · 2026-10-27
- Latent-Foresight: world models that learn to predict their own future — abursuc · 2026-10-27
- UK AISI papers: no good safeguards yet against misuse of biological AI tools — davidmanheim · 2026-10-27
- CMU's XYBench: LLMs answer queries literally and reinforce user misconceptions — valentina__py · 2026-10-27