# NeoHorse-1：通过路由框架的智能体后训练实现递归自我改进

- 来源：HuggingFace Daily Papers（社区热门论文）
- 发布时间：2026-09-08 08:00
- AIHOT 分数：40
- AIHOT 链接：https://aihot.news/items/cmttieljs0a4nrofp2ahn07h4
- 原文链接：https://arxiv.org/abs/2609.08183

## AI 摘要

NeoHorse-1 是一个智能体原生模型系列，通过结合异构模型池与智能路由，将用户交互记录转化为训练样本，并利用路由信号组织三阶段监督微调和路由引导的在线蒸馏，形成评估-选择-更新的闭环。

## 正文

Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabilities and converts that evidence into the next round of learning. We present NeoHorse-1, a family of agent-native models developed to explore this path through agentic post-training. Our system combines a heterogeneous model pool with intelligent routing, recording the predicted capability demand, selected service tier, and subsequent interaction for each user turn. These records are converted into training examples that preserve interleaved reasoning, tool calls, and harness context, and are admitted through structural validation, six-dimensional semantic evaluation, and subscene-level labeling. Routing signals organize supervised fine-tuning into a three-stage curriculum and extend to routing-guided on-policy distillation, where a teacher supervises student-generated responses under the same progression. Capability-guided allocation then converts evaluation feedback into the next training mixture, closing an evaluation-selection-update loop in which what the system learns to do shapes what it learns from next. Across eleven benchmarks covering harness-based agents, tool use, coding, and instruction following, post-training raises the macro-average from 58.94 to 64.87 at 4B and from 65.60 to 69.04 at 9B, substantially narrowing the aggregate gap between the post-trained 4B model and the 9B base model. NeoHorse-1 provides an initial prototype of this feedback-driven process and a path toward harness-mediated RSI across successive iterations.