# Reasoning Primitives in Hybrid and Non-Hybrid LLMs: Do Architectural Differences Yield Advantages in State-Tracking and Recall?

arXiv: https://arxiv.org/abs/2604.21454 · HTML: https://mlsanity.com/paper/2604.21454/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 14 of its 21 references that are in the map.
- First posted on arXiv 23 Apr 2026.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 14 of 14)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality](https://mlsanity.com/paper/2405.21060/) (2024)
- [GLM-5: from Vibe Coding to Agentic Engineering](https://mlsanity.com/paper/2602.15763/) (2026)
- [Gated Delta Networks: Improving Mamba2 with Delta Rule](https://mlsanity.com/paper/2412.06464/) (2024)
- [Quantifying the Carbon Emissions of Machine Learning](https://mlsanity.com/paper/1910.09700/) (2019)
- [Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach](https://mlsanity.com/paper/2502.05171/) (2025)
- [A Survey on Large Language Model Acceleration based on KV Cache Management](https://mlsanity.com/paper/2412.19442/) (2024)
- [Olmo 3](https://mlsanity.com/paper/2512.13961/) (2025)
- [Logical Reasoning in Large Language Models: A Survey](https://mlsanity.com/paper/2502.09100/) (2025)
- [Olmo Hybrid: From Theory to Practice and Back](https://mlsanity.com/paper/2604.03444/) (2026)
- [LongCoT: Benchmarking Long-Horizon Chain-of-Thought Reasoning](https://mlsanity.com/paper/2604.14140/) (2026)
- [State over Tokens: Characterizing the Role of Reasoning Tokens](https://mlsanity.com/paper/2512.12777/) (2025)
- [Thinking Augmented Pre-training](https://mlsanity.com/paper/2509.20186/) (2025)
- [The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures](https://mlsanity.com/paper/2510.05364/) (2025)

## Built on by (most cited first, 1 of 1)

- [The Capability Convergence Hypothesis: Capability from Access Structure, Not Scale](https://mlsanity.com/paper/2607.14144/) (2026)

## Abstract (arXiv)

Reasoning in large language models is often discussed as a single capability, but some of its gains may stem from simpler underlying operations. We examine two such primitives, recall and state-tracking, through five controlled task families centered on state-based recall, and compare matched transformer and hybrid architectures with and without reasoning augmentation. Across the suite, reasoning-augmented variants substantially outperform instruction-only variants, often by large margins. This pattern is consistent with the State over Tokens view: externalized reasoning traces help because they carry the intermediate state forward in token space. By contrast, hybrid inductive bias does not yield a uniform advantage in accuracy once reasoning tokens are available. When architectural differences do appear, they follow task structure: the hybrid Think model is more robust on strictly sequential chained updates, whereas the transformer Think model is more robust on flat multi-hop retrieval. We therefore cast the main contribution of this study as a descriptive account of what drives performance on state-based recall tasks: reasoning-token augmentation appears to be the dominant factor, while hybrid advantages are narrower, task-dependent, and potentially more about inference efficiency than overall capability. We also release the codebase and data required to reproduce these results.