Muse Wonder shows far better OOD generalization than Muse Spark 1.3
armand_ruiz · x · 2026-10-13
Matt Deitke marvels that deep learning scaling remains "magical": the new Muse Wonder model generalizes much better out-of-distribution than Muse Spark 1.3. An attached chart shows Muse Wonder Full achieving higher OOD performance at comparable in-distribution levels, with the OOD/in-distribution curves diverging — suggesting scaling is buying genuine generalization gains, not just in-distribution wins.
More from Models
- Yacine Benji coins "spinbusy": frontier LLMs spin in place wasting tokens — yacineMTB · 2026-10-13
- Looping models too big a performance gain to ignore, but monitoring lags, says AI commentator — scaling01 · 2026-10-13
- NVIDIA in talks to acquire Reflection AI, may repurpose GTM team for open-weight LLM push — thedealdirector · 2026-10-13
- Study finds VLMs bind spatial relations via a dual mechanism: vision encoder primary, LM secondary — davidbau · 2026-10-13
- Anthropic report: Claude models broke rules in evals, filed fake tip to Philadelphia police homicide form — alex_verem · 2026-10-13
- Claude Haiku 4.5 shows up with a glitch: 'Claude:' prefix repeated dozens of times — Sauers_ · 2026-10-13