We introduce Native Parallel Reasoner (NPR), a teacher-free framework that enables Large Language Models (LLMs) to self-evolve genuine parallel reasoning capabilities. NPR transforms the model from sequential emulation to native parallel cognition through three key innovations: 1) a self-distilled progressive training paradigm that transitions from ``cold-start'' format discovery to strict topological constraints without external supervision; 2) a novel Parallel-Aware Policy Optimization (PAPO) algorithm that optimizes branching policies directly within the execution graph, allowing the model to learn adaptive decomposition via trial and error; and 3) a robust NPR Engine that refactors memory management and flow control of SGLang to enable stable, large-scale parallel RL training. Across eight reasoning benchmarks, NPR trained on Qwen3-4B achieves performance gains of up to 24.5% and inference speedups up to 4.6x. Unlike prior baselines that often fall back to autoregressive decoding, NPR demonstrates 100% genuine parallel execution, establishing a new standard for self-evolving, efficient, and scalable agentic reasoning.

翻译：本文提出原生并行推理器（Native Parallel Reasoner，NPR），一种无需教师模型的框架，使大型语言模型（LLMs）能够自我演化出真正的并行推理能力。NPR通过三项关键创新将模型从序列化模拟转变为原生并行认知：1）一种自蒸馏渐进式训练范式，在无外部监督的情况下从“冷启动”格式发现过渡到严格的拓扑约束；2）一种新颖的并行感知策略优化（Parallel-Aware Policy Optimization，PAPO）算法，直接在执行图中优化分支策略，使模型能够通过试错学习自适应分解；3）一个稳健的NPR引擎，重构了SGLang的内存管理与流程控制，以实现稳定的大规模并行强化学习训练。在八个推理基准测试中，基于Qwen3-4B训练的NPR实现了高达24.5%的性能提升和高达4.6倍的推理加速。与先前常退化为自回归解码的基线方法不同，NPR展示了100%的真正并行执行，为自我演化、高效且可扩展的智能体推理设立了新标准。