Latent reasoning
Technique · Architecture & Efficiency · 3 items since Jun 2026
Models reason in hidden states rather than writing out every step as text.
Trends
Items
- Oct 9
LLoCoT replaces chain-of-thought with latent loops
A looped latent-reasoning framework replaces left-to-right chain-of-thought with iterative refinement of a fixed latent workspace, matching reasoning-SFT accuracy on code benchmarks while cutting time-to-first-answer-token ~36x and reasoning-phase latency ~42x.
- Jun 11
Switch makes latent reasoning RL-trainable
A pair of explicit boundary tokens makes hidden-state-recurrence latent chain-of-thought trainable with standard GRPO and open to mechanistic probing, reaching 79.3% on MATH-500 and showing the latent step does causally necessary computation.
- Jun 3
ALAR escalates from latent to explicit reasoning
Adaptive Latent Agentic Reasoning defaults to compact latent reasoning on routine agent turns and escalates to explicit chain-of-thought only for harder decisions, preserving accuracy while cutting generated tokens by up to 43.6% on agentic search and 84.6% on tool use.