# Mixture of Reasonings: Teach Large Language Models to Reason with Adaptive Strategies

arXiv: https://arxiv.org/abs/2507.00606 · HTML: https://mlsanity.com/paper/2507.00606/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Prompting and chain-of-thought reasoning, one of 17 directions in LLM reasoning.
- Builds on 19 of its 20 references that are in the map.
- First posted on arXiv 1 Jul 2025.

## Placed in

- LLM reasoning / [Prompting and chain-of-thought reasoning](https://mlsanity.com/lr/prompting-and-chain-of-thought-reasoning/)

## Builds on (most cited first, 19 of 19)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering](https://mlsanity.com/paper/1809.09600/) (2018)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategies](https://mlsanity.com/paper/2101.02235/) (2021)
- [Instruction Tuning for Large Language Models: A Survey](https://mlsanity.com/paper/2308.10792/) (2023)
- [Fine-Tuning Language Models with Just Forward Passes](https://mlsanity.com/paper/2305.17333/) (2023)
- [Unleashing the Emergent Cognitive Synergy in Large Language Models: A Task-Solving Agent through Multi-Persona Self-Collaboration](https://mlsanity.com/paper/2307.05300/) (2023)
- [Take a Step Back: Evoking Reasoning via Abstraction in Large Language Models](https://mlsanity.com/paper/2310.06117/) (2023)
- [Self-Discover: Large Language Models Self-Compose Reasoning Structures](https://mlsanity.com/paper/2402.03620/) (2024)
- [Large Language Models as Analogical Reasoners](https://mlsanity.com/paper/2310.01714/) (2023)
- [Think Twice: Perspective-Taking Improves Large Language Models' Theory-of-Mind Capabilities](https://mlsanity.com/paper/2311.10227/) (2023)
- [Meta Reasoning for Large Language Models](https://mlsanity.com/paper/2406.11698/) (2024)
- [AutoPRM: Automating Procedural Supervision for Multi-Step Reasoning via Controllable Question Decomposition](https://mlsanity.com/paper/2402.11452/) (2024)
- [Fast Trainable Projection for Robust Fine-Tuning](https://mlsanity.com/paper/2310.19182/) (2023)
- [DolphCoder: Echo-Locating Code Large Language Models with Diverse and Multi-Objective Instruction Tuning](https://mlsanity.com/paper/2402.09136/) (2024)

## Built on by (most cited first, 3 of 3)

- [Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey](https://mlsanity.com/paper/2507.09662/) (2025)
- [From Efficiency to Adaptivity: A Deeper Look at Adaptive Reasoning in Large Language Models](https://mlsanity.com/paper/2511.10788/) (2025)
- [Step Potential Advantage Estimation: Harnessing Intermediate Confidence and Correctness for Efficient Mathematical Reasoning](https://mlsanity.com/paper/2601.03823/) (2026)

## Abstract (arXiv)

Large language models (LLMs) excel in complex tasks through advanced prompting techniques like Chain-of-Thought (CoT) and Tree-of-Thought (ToT), but their reliance on manually crafted, task-specific prompts limits adaptability and efficiency. We introduce Mixture of Reasoning (MoR), a training framework that embeds diverse reasoning strategies into LLMs for autonomous, task-adaptive reasoning without external prompt engineering. MoR has two phases: Thought Generation, creating reasoning chain templates with models like GPT-4o, and SFT Dataset Construction, pairing templates with benchmark datasets for supervised fine-tuning. Our experiments show that MoR significantly enhances performance, with MoR150 achieving 0.730 (2.2% improvement) using CoT prompting and 0.734 (13.5% improvement) compared to baselines. MoR eliminates the need for task-specific prompts, offering a generalizable solution for robust reasoning across diverse tasks.