2026-11-17 · America/Los_Angeles · 论文 · #14
Energy-Based Reasoners: Your Recurrent Reasoner Should Be an Energy-Based Model
Looping is a powerful inductive bias for reasoning. By repeatedly iterating a recurrent block, looped models can scale their effective depth and enable multi-step reasoning in latent space. However, existing looped models are not inherently guided by an explicit measure of progress, making it difficult to determine when to stop looping and how to improve upon the current state. To enable this, we introduce Energy-Based Reasoners (EBRs), which learn an energy function over latent reasoning states and formulate looped updates as gradient descent on this energy. The energy scores latent states,…
热度 40.9 / 100;排名与评分保留该期记录。