AERO: Autonomous Evolutionary Reasoning Optimization via Endogenous Dual-Loop Feedback

TL;DR

AERO is an unsupervised framework that enhances large language model reasoning by internalizing self-questioning, answering, and criticism through a dual-loop system inspired by ZPD, leading to improved performance across multiple benchmarks.

Contribution

It introduces AERO, a novel autonomous reasoning optimization method with entropy-based positioning, counterfactual correction, and staggered training to improve LLM reasoning without external supervision.

Findings

Achieves average performance improvements of 4.57% and 5.10% on two benchmarks.

Outperforms competitive baselines across nine diverse benchmarks.

Demonstrates effective autonomous reasoning evolution in LLMs.

Abstract

Large Language Models (LLMs) have achieved significant success in complex reasoning but remain bottlenecked by reliance on expert-annotated data and external verifiers. While existing self-evolution paradigms aim to bypass these constraints, they often fail to identify the optimal learning zone and risk reinforcing collective hallucinations and incorrect priors through flawed internal feedback. To address these challenges, we propose \underline{A}utonomous \underline{E}volutionary \underline{R}easoning \underline{O}ptimization (AERO), an unsupervised framework that achieves autonomous reasoning evolution by internalizing self-questioning, answering, and criticism within a synergistic dual-loop system. Inspired by the \textit{Zone of Proximal Development (ZPD)} theory, AERO utilizes entropy-based positioning to target the ``solvability gap'' and employs Independent Counterfactual Correction for robust verification. Furthermore, we introduce a Staggered Training Strategy to synchronize capability growth across functional roles and prevent curriculum collapse. Extensive evaluations across nine benchmarks spanning three domains demonstrate that AERO achieves average performance improvements of 4.57\% on Qwen3-4B-Base and 5.10\% on Qwen3-8B-Base, outperforming competitive baselines. Code is available at https://github.com/mira-ai-lab/AERO.