---
id: 20260704-T0-03
title: "超越下一个词预测：RLVR让LLM学会正确调用Atlassian API"
title_en: "Beyond Next-Token: RLVR Trains LLMs to Use Atlassian APIs Correctly"
url: https://ai.daily.yangsir.net/daily/20260704-T0-03
issue_date: 2026-07-04
publish_date: 2026-07-03T04:00:00.000Z
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2607.01465
---

# 超越下一个词预测：RLVR让LLM学会正确调用Atlassian API

LLM被训练预测下一个词，而非在具体API中正确执行操作。该研究提出基于RLVR的方法，在Atlassian工作流上训练LLM学会按顺序、带嵌套参数地正确调用API端点。相比GPT-4，该方法在复杂SaaS工作流上的成功率提升了40%以上。

## English Version

**Beyond Next-Token: RLVR Trains LLMs to Use Atlassian APIs Correctly**

LLMs are trained for next-token prediction, not for correct API execution. This paper uses RLVR to train LLMs on Atlassian workflows, achieving over 40% higher success rate than GPT-4 in correctly calling endpoints with nested arguments in order.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2607.01465)

**详情页**：https://ai.daily.yangsir.net/daily/20260704-T0-03

---

*智语观潮 · Daily — https://ai.daily.yangsir.net/llms.txt*