BDH-CQ: IN-CONTEXT LEARNING WITH RECURRENT LATENT REASONING [R]
Heat trend
Collecting trend data
The percentage is based on available heat signal, not comment count or independent people.
BDH-CQ is a reasoning system that integrates memory, adaptation, and inference. It updates recurrent memory with demonstrations of new tasks and solves queries through iterative co…
We introduce BDH-CQ, a reasoning system that brings these capabilities together. Demonstrations of a previously unseen task update recurrent memory; the query is then solved through iterative computation in a high-dimensional latent workspace. Intermediate reasoning states are not decoded into language. BDH-CQ makes memory, adaptation, and inference part of the same computational fabric. Inputs presented at inference time continuously update the model’s recurrent memory; the model then solves a query through iterative computation in a high-dimensional latent space, without verbalizing its intermediate reasoning. Neither task identifiers nor evaluation-task demonstration pairs participate in training, and no parameters are updated at inference time. A 150M-parameter configuration reaches 29.5% pass@2 on ARC-AGI-1 at a computed $0.00070 per task, breaking through the previously reported cost–accuracy Pareto frontier.