# LLMs Improving LLMs: Agentic Discovery for Test-Time Scaling

arXiv: https://arxiv.org/abs/2605.08083 · HTML: https://mlsanity.com/paper/2605.08083/

## Key facts

- Cited by 10 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 40 of its 46 references that are in the map.
- First posted on arXiv 8 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, 30 of 40)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [Neural Architecture Search with Reinforcement Learning](https://mlsanity.com/paper/1611.01578/) (2016)
- [s1: Simple test-time scaling](https://mlsanity.com/paper/2501.19393/) (2025)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations](https://mlsanity.com/paper/2312.08935/) (2023)
- [AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://mlsanity.com/paper/2506.13131/) (2025)
- [Improve Mathematical Reasoning in Language Models by Automated Process Supervision](https://mlsanity.com/paper/2406.06592/) (2024)
- [Automated Design of Agentic Systems](https://mlsanity.com/paper/2408.08435/) (2024)
- [Meta-Harness: End-to-End Optimization of Model Harnesses](https://mlsanity.com/paper/2603.28052/) (2026)
- [Evolution of Heuristics: Towards Efficient Automatic Algorithm Design Using Large Language Model](https://mlsanity.com/paper/2401.02051/) (2024)
- [Dynamic Early Exit in Reasoning Models](https://mlsanity.com/paper/2504.15895/) (2025)
- [Deep Think with Confidence](https://mlsanity.com/paper/2508.15260/) (2025)
- [Reasoning Models Know When They're Right: Probing Hidden States for Self-Verification](https://mlsanity.com/paper/2504.05419/) (2025)
- [Let's Sample Step by Step: Adaptive-Consistency for Efficient Reasoning and Coding with LLMs](https://mlsanity.com/paper/2305.11860/) (2023)
- [Learning to Discover at Test Time](https://mlsanity.com/paper/2601.16175/) (2026)
- [Confidence Improves Self-Consistency in LLMs](https://mlsanity.com/paper/2502.06233/) (2025)
- [Escape Sky-high Cost: Early-stopping Self-Consistency for Multi-step Reasoning](https://mlsanity.com/paper/2401.10480/) (2024)
- [Efficient Test-Time Scaling via Self-Calibration](https://mlsanity.com/paper/2503.00031/) (2025)
- [ThetaEvolve: Test-time Learning on Open Problems](https://mlsanity.com/paper/2511.23473/) (2025)
- [Make Every Penny Count: Difficulty-Adaptive Self-Consistency for Cost-Efficient Reasoning](https://mlsanity.com/paper/2408.13457/) (2024)
- [Reasoning Aware Self-Consistency: Leveraging Reasoning Paths for Efficient LLM Sampling](https://mlsanity.com/paper/2408.17017/) (2024)
- [Sampling-Efficient Test-Time Scaling: Self-Estimating the Best-of-N Sampling in Early Decoding](https://mlsanity.com/paper/2503.01422/) (2025)
- [Wider or Deeper? Scaling LLM Inference-Time Compute with Adaptive Branching Tree Search](https://mlsanity.com/paper/2503.04412/) (2025)
- [The Majority is not always right: RL training for solution aggregation](https://mlsanity.com/paper/2509.06870/) (2025)
- [Answer Convergence as a Signal for Early Stopping in Reasoning](https://mlsanity.com/paper/2506.02536/) (2025)
- [ParaThinker: Native Parallel Thinking as a New Paradigm to Scale LLM Test-time Compute](https://mlsanity.com/paper/2509.04475/) (2025)

## Built on by (most cited first, 10 of 10)

- [Dream-RSI: Recursive Self-Improvement through Evolving Worlds](https://mlsanity.com/paper/2609.14858/) (2026)
- [AI for Auto-Research: Roadmap & User Guide](https://mlsanity.com/paper/2605.18661/) (2026)
- [Share More, Search Less: Collaborative Parallel Thinking for Efficient Test-Time Scaling](https://mlsanity.com/paper/2605.27030/) (2026)
- [Small RL Controller, Large Language Model: RL-Guided Adaptive Sampling for Test-Time Scaling](https://mlsanity.com/paper/2606.03102/) (2026)
- [What is Missing from AI Post-Training AI: An Empirical Analysis](https://mlsanity.com/paper/2608.19072/) (2026)
- [LatentSift: Policy-State Filtering for Token-Efficient Verification of Software Engineering Agents](https://mlsanity.com/paper/2609.36371/) (2026)
- [Align Then Reason: A Multimodal Lip-Sync Judge for Dubbing](https://mlsanity.com/paper/2610.00825/) (2026)
- [CARE: Certifying Acceleration for Vision-Language-Action Inference](https://mlsanity.com/paper/2610.08917/) (2026)
- [From Pareto to Preference: Personalized Test-Time Scaling via Amortized Agentic Policy Discovery](https://mlsanity.com/paper/2610.09684/) (2026)
- [From Existing Methods to New Research Ideas: Two Perspectives on Recursive Self-Improvement](https://mlsanity.com/paper/2610.14625/) (2026)

## Abstract (arXiv)

Test-time scaling (TTS) has become an effective approach for improving large language model performance by allocating additional computation during inference. However, existing TTS strategies are largely hand-crafted: researchers manually design reasoning patterns and tune heuristics by intuition, leaving much of the computation-allocation space unexplored. We propose an environment-driven framework, AutoTTS, that changes what researchers design: from individual TTS heuristics to environments where TTS strategies can be discovered automatically. The key to AutoTTS lies in environment construction: the discovery environment must make the control space tractable and provide cheap, frequent feedback for TTS search. As a concrete instantiation, we formulate width--depth TTS as controller synthesis over pre-collected reasoning trajectories and probe signals, where controllers decide when to branch, continue, probe, prune, or stop and can be evaluated cheaply without repeated LLM calls. We further introduce beta parameterization to make the search tractable and fine-grained execution trace feedback to improve discovery efficiency by helping the agent diagnose why a TTS program fails. Experiments on mathematical reasoning benchmarks show that the discovered strategies improve the overall accuracy--cost tradeoff over strong manually designed baselines. The discovered strategies generalize to held-out benchmarks and model scales, while the entire discovery costs only 39.9 and 160 minutes. Our data, and code will be open-source at https://github.com/zhengkid/AutoTTS.