Paper Author: The Stronger the Model, the Better Geometry-Based Failure Prediction Works
nsaphra · x · 2026-11-07
Naomi Saphra adds the coolest finding from her Adversarial Concept Search paper update: at every scale checked, the better the model (and the better it is at a task), the better the geometry-based method works at finding its failure modes—meaning the approach gains rather than loses utility as models improve.
More from Research
- Q-OIL, an undergrad first paper at CoRL2026, treats human interventions as soft exploration guidance — aidenybai · 2026-11-07
- Andrew Carr reignites the MoE debate: do tokens choose experts or experts choose tokens? — andrew_n_carr · 2026-11-07
- AdderBoard challenge: smallest transformer that adds two 10-digit numbers, records keep falling — DimitrisPapail · 2026-11-07
- Zellige AI open-sources Tessera-27B-B-CJK-RL: RL-boosted CJK-thinking coder cuts thinking tokens by a third, matches base on hard math — ghgi_ · 2026-11-07
- Unverified claim: frontier labs skewing ICLR reviews with malicious reviews, scandal brewing — examachine · 2026-11-07
- Persona Atlas visualizes Qwen's experts and heads, exposing bizarre inner personas — Sauers_ · 2026-11-07