mlsanityData to 14 Oct 2026

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

arXiv 2610.16491 ↗First posted 12 Oct 2026cs.LG cs.AIcited by 0

On the map: LLM agents and tool use → Software engineering coding agents

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.

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.

Builds on

26 of 39 references.

Closest ancestors · 3

Same direction, newest first

  1. 2025Huxley-G"odel Machine: Human-Level Coding Agent Development by an Approximation of the Optimal Self-Improving Machine24 Oct
  2. 2025A Self-Improving Coding Agent21 Apr
  3. 2024AFlow: Automating Agentic Workflow Generation14 Oct

Other references · 23

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
  3. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  4. 2023Let's Verify Step by Stepcited by 3,789
  5. 2023Tree of Thoughts: Deliberate Problem Solving with Large Language Modelscited by 3,211
  6. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,653
  7. 2025GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learningcited by 542
  8. 2024Automated Design of Agentic Systemscited by 407
  9. 2024EAGLE-2: Faster Inference of Language Models with Dynamic Draft Treescited by 370
  10. 2023SpecInfer: Accelerating Generative Large Language Model Serving with Tree-based Speculative Inference and Verificationcited by 365
  11. 2024Better & Faster Large Language Models via Multi-token Predictioncited by 360
  12. 2025Darwin Godel Machine: Open-Ended Evolution of Self-Improving Agentscited by 356
11 more
  1. 2024ArCHer: Training Language Model Agents via Hierarchical Multi-Turn RLcited by 228
  2. 2025MemEvolve: Meta-Evolution of Agent Memory Systemscited by 95
  3. 2026Hyperagentscited by 56
  4. 2025Dynamic Speculative Agent Planningcited by 12
  5. 2013Approximation Algorithms for Stochastic Boolean Function Evaluation and Stochastic Submodular Set Covercited by 10
  6. 2025Bridging Draft Policy Misalignment: Group Tree Optimization for Speculative Decodingcited by 10
  7. 2025Draft, Verify, and Improve: Toward Training-Aware Speculative Decodingcited by 4
  8. 2026FlashEvolve: Accelerating Agent Self-Evolution with Asynchronous Stage Orchestrationcited by 3
  9. 2026Janus: An Algorithm-Evaluator Co-Evolution Framework for LLM-Driven Discovery under Expensive Evaluation Budgetscited by 3
  10. 2022Adaptivity Gaps for the Stochastic Boolean Function Evaluation Problemcited by 1
  11. 2026Stochastic Function Certification with Correlationscited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026SpecForge: Recursive Self-Improvement for Coding Agents via Behavioral Specifications
  2. 2026RRSI: Regularized Recursive Self-Improvement of Agent Harnesses
  3. 2026MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution
  4. 2026RSIGym: A Flexible Environment for Recursive Self-Improvement
  5. 2026Metan: Recursive Self-Improvement through Emergent Depth
  6. 2026The Red Queen G"odel Machine: Co-Evolving Agents and Their Evaluators
  7. 2026Reward-Free Evolving Agents via Pairwise Validator
  8. 2026Verification and Self-Improvement in Agentic AI: Foundations and Limits