# Drafting Agents, Not Just Tokens: Speculative Recursive Self-Improvement for Agentic Systems

arXiv: https://arxiv.org/abs/2610.16491 · HTML: https://mlsanity.com/paper/2610.16491/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 26 of its 39 references that are in the map.
- First posted on arXiv 12 Oct 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 26 of 26)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning](https://mlsanity.com/paper/2507.19457/) (2025)
- [AFlow: Automating Agentic Workflow Generation](https://mlsanity.com/paper/2410.10762/) (2024)
- [Automated Design of Agentic Systems](https://mlsanity.com/paper/2408.08435/) (2024)
- [EAGLE-2: Faster Inference of Language Models with Dynamic Draft Trees](https://mlsanity.com/paper/2406.16858/) (2024)
- [SpecInfer: Accelerating Generative Large Language Model Serving with Tree-based Speculative Inference and Verification](https://mlsanity.com/paper/2305.09781/) (2023)
- [Better & Faster Large Language Models via Multi-token Prediction](https://mlsanity.com/paper/2404.19737/) (2024)
- [Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agents](https://mlsanity.com/paper/2505.22954/) (2025)
- [ArCHer: Training Language Model Agents via Hierarchical Multi-Turn RL](https://mlsanity.com/paper/2402.19446/) (2024)
- [MemEvolve: Meta-Evolution of Agent Memory Systems](https://mlsanity.com/paper/2512.18746/) (2025)
- [A Self-Improving Coding Agent](https://mlsanity.com/paper/2504.15228/) (2025)
- [Hyperagents](https://mlsanity.com/paper/2603.19461/) (2026)
- [Huxley-G"odel Machine: Human-Level Coding Agent Development by an Approximation of the Optimal Self-Improving Machine](https://mlsanity.com/paper/2510.21614/) (2025)
- [Dynamic Speculative Agent Planning](https://mlsanity.com/paper/2509.01920/) (2025)
- [Approximation Algorithms for Stochastic Boolean Function Evaluation and Stochastic Submodular Set Cover](https://mlsanity.com/paper/1303.0726/) (2013)
- [Bridging Draft Policy Misalignment: Group Tree Optimization for Speculative Decoding](https://mlsanity.com/paper/2509.22134/) (2025)
- [Draft, Verify, and Improve: Toward Training-Aware Speculative Decoding](https://mlsanity.com/paper/2510.05421/) (2025)
- [FlashEvolve: Accelerating Agent Self-Evolution with Asynchronous Stage Orchestration](https://mlsanity.com/paper/2605.08520/) (2026)
- [Janus: An Algorithm-Evaluator Co-Evolution Framework for LLM-Driven Discovery under Expensive Evaluation Budgets](https://mlsanity.com/paper/2608.08189/) (2026)
- [Adaptivity Gaps for the Stochastic Boolean Function Evaluation Problem](https://mlsanity.com/paper/2208.03810/) (2022)
- [Stochastic Function Certification with Correlations](https://mlsanity.com/paper/2604.02611/) (2026)

## Abstract (arXiv)

Improving an agent through successive revisions requires balancing exploration and evaluation under a limited budget. A narrow search can miss useful sequences of edits, while fully evaluating every intermediate version can exhaust the budget before those sequences yield a verified improvement. We introduce Speculative Recursive Self-Improvement (SpecRSI), which drafts executable successor agents before their predecessors are fully evaluated, while requiring complete verification before adoption. Verification-Aware Tree Search (VATS) allocates the budget between further edits and outstanding checks using the predicted verified return of a continuation. Group Verification Optimization (GVO) learns the improver from completed searches while keeping the real evaluator fixed. In the main scientific comparison, terminal gains exceed Janus by 33.3%--40.0%. Additional evaluations show that SpecRSI outperforms Janus in all 50 scientific backbone--task pairs and exceeds MetaRSI-v1 by 1.85%--2.38% in macro-average system score across twenty backbones. Coding-validation success reaches 69.4% versus HGM's 64.3% at a 20.8% lower target-cost index.