Entropy doesn't track what matters in chains of thought, BlackboxNLP paper finds
gsarti_ · x · 2026-10-29
A paper by Sara Candussio's team, accepted to the BlackboxNLP reproducibility track, asks whether entropy in a chain of thought actually tracks what's important. The answer: mostly no. The authors also identify the spurious cause behind why it previously looked like it did. The work is being presented at BlackboxNLP (16:00-17:00), with a detailed thread available.
More from Research
- Weaviate Podcast: MIT PhD's "Machine Studying" lets always-on agents prep without knowing the task — CShorten30 · 2026-10-29
- The singularity's week: 80-year-old cells reset, 25-year-old math disproved by AI labs — rand_longevity · 2026-10-29
- Single-nucleus RNA-seq atlas reveals distinct features of early-onset colorectal cancer in the young — EricTopol · 2026-10-29
- Stanford's ChromBPNet Reads DNA Regulatory Switches Base by Base — RichardsonDx · 2026-10-29
- Determinism in Training Costs Single-Digit Percent, Not 10x — gandamu_ml · 2026-10-29
- Neuron steering struggles at persona creation, failing multi-turn stability — davidad · 2026-10-29