Beyond Memorization: Extending Reasoning Depth with Recurrence, Memory and Test-Time Compute Scaling
This work addresses the problem of improving multi-step reasoning in AI models, which is crucial for advancing their cognitive abilities, though it appears incremental as it builds on existing architectures with specific enhancements.
The study investigated how neural architectures and training methods affect multi-step reasoning in large language models using a cellular automata framework, finding that while models achieve high accuracy in next-state prediction, their performance declines sharply for multi-step reasoning, but extending effective model depth with recurrence, memory, and test-time compute scaling substantially enhances these capabilities.
Reasoning is a core capability of large language models, yet understanding how they learn and perform multi-step reasoning remains an open problem. In this study, we explore how different architectures and training methods affect model multi-step reasoning capabilities within a cellular automata framework. By training on state sequences generated with random Boolean functions for random initial conditions to exclude memorization, we demonstrate that most neural architectures learn to abstract the underlying rules. While models achieve high accuracy in next-state prediction, their performance declines sharply if multi-step reasoning is required. We confirm that increasing model depth plays a crucial role for sequential computations. We demonstrate that an extension of the effective model depth with recurrence, memory, and test-time compute scaling substantially enhances reasoning capabilities.