Ask about a blue sphere and the logits leak its material and size
sineadwilliamso · x · 2026-10-07
Key findings from the paper: the residual stream encodes nearly everything about a scene, queried or not; even the top 2 logits leak task-irrelevant info — ask 'is there a blue sphere?' and the logits reveal the sphere's material and size, never mentioned in the query.
More from Research
- GPU-Free Activation Alignment Recovers Half of Full-Context Performance for Tabular ICL — Independent-Researcher · 2026-10-07
- OPPD Distills Power-Distribution Sampling Into One Generation, Beating 64-Candidate Sampling — USC · 2026-10-07
- Deleting Definitions Changes Nothing: Typed Decision Models Follow Labels, Not Rules — USC · 2026-10-07
- Auxiliary loss forces hybrid LMs like Qwen3.5 to actually use recurrent memory, +12.1% on agentic tasks — mohitban47 · 2026-10-07
- COLM paper: latent reasoning traces decodable 65-93% of the time in LRMs — sarahwiegreffe · 2026-10-07
- Apple ML Research is hiring a PhD intern to make foundation models know what they know — sineadwilliamso · 2026-10-07