Machine Heart
May 18, 2026 · Artificial Intelligence
Can Large Models Reason Deeply with Only a Few Thinking Tokens?
The paper introduces Heima, a framework that compresses chain‑of‑thought reasoning into a small set of abstract “thinking tokens” for multimodal large models, dramatically reducing generated tokens while preserving inference capability, and provides an adaptive interpreter to reconstruct human‑readable reasoning for analysis.
Efficient Inferencechain-of-thoughtlatent reasoning
0 likes · 12 min read