Lesioned generative models hurt human mass discrimination only across surfaces
JoshHMcDermott · x · 2026-11-10
Part 10 of McDermott's thread: the team manipulated whether surface resonances conformed to real-world object statistics by sampling from intact or 'lesioned' generative models. Lesions impaired humans' ability to discriminate ball mass, but only when the two surfaces differed (e.g., ceramic vs. plastic), showing reliance on learned acoustic priors depends on the inference scenario.
More from Research
- Google DeepMind and Artificial Analysis launch MKA benchmark; Claude Fable 5.2 tops at 57% Climb WSR — ArtificialAnlys · 2026-11-11
- MBZUAI releases open world model PAN 2: LLM planning plus 29B diffusion hits 75.1% step accuracy — rohanpaul_ai · 2026-11-11
- ProtoMotions Deployment Stack: One Exported Graph, 1.3 ms C++ Loop for Real Robots — carlosdponx · 2026-11-11
- MKA deep dive: top models cluster at 95-97% on basics but split 28-57% on hard tasks — ArtificialAnlys · 2026-11-11
- MKA Benchmark by Output Type: No Model Tops 52% on Word or Text Files — ArtificialAnlys · 2026-11-11
- DeepMind-backed MKA benchmark crowns Claude Fable 5.2 at 57%, with 30x cost spread — ArtificialAnlys · 2026-11-11