# M2A: Synergizing Mathematical and Agentic Reasoning in Large Language Models arXiv: https://arxiv.org/abs/2605.09879 ยท HTML: https://mlsanity.com/paper/2605.09879/ ## Key facts - Cited by 1 papers in the map (data through 14 Oct 2026). - Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning. - Builds on 28 of its 50 references that are in the map. - First posted on arXiv 11 May 2026. ## Placed in - LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/) - LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/) ## Builds on (most cited first, 28 of 28) - [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025) - [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025) - [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021) - [Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities](https://mlsanity.com/paper/2507.06261/) (2025) - [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023) - [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024) - [MathArena: Evaluating LLMs on Uncontaminated Math Competitions](https://mlsanity.com/paper/2505.23281/) (2025) - [SWE-smith: Scaling Data for Software Engineering Agents](https://mlsanity.com/paper/2504.21798/) (2025) - [AdaMerging: Adaptive Model Merging for Multi-Task Learning](https://mlsanity.com/paper/2310.02575/) (2023) - [AlphaEdit: Null-Space Constrained Knowledge Editing for Language Models](https://mlsanity.com/paper/2410.02355/) (2024) - [Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains](https://mlsanity.com/paper/2503.23829/) (2025) - [General-Reasoner: Advancing LLM Reasoning Across All Domains](https://mlsanity.com/paper/2505.14652/) (2025) - [Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning](https://mlsanity.com/paper/2507.00432/) (2025) - [A Survey of Agentic Reasoning for Large Language Models: Towards Recursively Self-Improving and Collective Agents](https://mlsanity.com/paper/2601.12538/) (2026) - [Kimi-Dev: Agentless Training as Skill Prior for SWE-Agents](https://mlsanity.com/paper/2509.23045/) (2025) - [SWE-Mirror: Scaling Issue-Resolving Datasets by Mirroring Issues Across Repositories](https://mlsanity.com/paper/2509.08724/) (2025) - [SWE-Lego: Pushing the Limits of Supervised Fine-tuning for Software Issue Resolving](https://mlsanity.com/paper/2601.01426/) (2026) - [SWE-World: Building Software Engineering Agents in Docker-Free Environments](https://mlsanity.com/paper/2602.03419/) (2026) - [Decouple and Orthogonalize: A Data-Free Framework for LoRA Merging](https://mlsanity.com/paper/2505.15875/) (2025) - [Reasoning and Tool-use Compete in Agentic RL:From Quantifying Interference to Disentangled Tuning](https://mlsanity.com/paper/2602.00994/) (2026) - [Multi-Task GRPO: Reliable LLM Reasoning Across Tasks](https://mlsanity.com/paper/2602.05547/) (2026) - [To Mix or To Merge: Toward Multi-Domain Reinforcement Learning for Large Language Models](https://mlsanity.com/paper/2602.12566/) (2026) - [SERA: Soft-Verified Efficient Repository Agents](https://mlsanity.com/paper/2601.20789/) (2026) - [RAIN-Merging: A Gradient-Free Method to Enhance Instruction Following in Large Reasoning Models with Preserved Thinking Format](https://mlsanity.com/paper/2602.22538/) (2026) - [Reasoning Curriculum: Bootstrapping Broad LLM Reasoning from Math](https://mlsanity.com/paper/2510.26143/) (2025) - [How and Why LLMs Generalize: A Fine-Grained Analysis of LLM Reasoning from Cognitive Behaviors to Low-Level Patterns](https://mlsanity.com/paper/2512.24063/) (2025) - [SWE-AGILE: A Software Agent Framework for Efficiently Managing Dynamic Reasoning Context](https://mlsanity.com/paper/2604.11716/) (2026) - [Orthogonal Model Merging](https://mlsanity.com/paper/2602.05943/) (2026) ## Built on by (most cited first, 1 of 1) - [BACAM: Behavior-Aware Continual Agent Merging for Multi-Turn Interaction](https://mlsanity.com/paper/2610.04966/) (2026) ## Abstract (arXiv) While reasoning has become a central capability of large language models (LLMs), the reasoning patterns required for different scenarios are often misaligned. Mathematical reasoning typically relies on intrinsic logic to solve closed-world problems in a single response, whereas agentic reasoning requires not only internal reasoning but also multi-turn interaction with external environments, interleaving thought and action. This misalignment prevents mathematical and agentic reasoning from effectively benefiting from each other, often yielding unstable reasoning behavior and only limited performance gains under multi-task learning. In this paper, we propose M2A, a novel paradigm that synergizes mathematical and agentic reasoning via model merging. To avoid overfitting to superficial reasoning patterns under joint training, M2A operates directly in parameter space: it identifies the feature subspace critical for agent behavior, and merges the mathematical reasoning task vector only along its null space, thereby injecting reasoning capability along directions that do not perturb agent behavior. Unlike SFT or RL, M2A requires no additional gradient-update and exposes the merging coefficient as a simple knob for controlling reasoning length. Experiments in a challenging real-world coding agent setting show that our method effectively extends agentic reasoning depth and delivers substantial performance improvements. Applied to a fine-tuned Qwen3-8B, M2A improves its SWE-Bench Verified resolved rate from 44.0% to 51.2% without retraining the model. Code is available at https://github.com/laplucky/M2A.git.