Jeff Dean spotlights analysis paper revealing hidden structures in word embeddings
niloofar_mire · x · 2026-11-22
Jeff Dean says he only discovered the paper (originally posted in February) via Martin Casado's tweet, praising the analysis and figures and the work by Dhruva Karkada, Daniel Korchinski, Andres Nava, Matthieu Wyart, and Yasaman Bahri for revealing "amazing hidden structures in word embeddings."
The thread is also a reminder that strong research often gets its audience via social rediscovery months after publication.
More from Research
- Bill Gates: speech rhythm shifts may offer an early clue to Alzheimer's risk — tallmetommy · 2026-11-22
- Viral French explainer breaks down the Riemann hypothesis proof without the jargon — ZeroStateReflex · 2026-11-22
- DeepMind bakes speculative decoding into Gemma 4: up to 3.1x server speedups, 2.2x on smartphone GPUs — dair_ai · 2026-11-22
- AI Is Clearing Open Erdős Problems: What It Says About Research Taste and AGI — gleech · 2026-11-22
- LlamaIndex's ParseBench-2 ranks 59 document parsers; KDL tops open-weight at 78.89 — victormustar · 2026-11-22
- When no human can follow machine-checked proofs, compression becomes scarce work — l4rz · 2026-11-22