---
id: 20261101-T0-07
title: "Seq@K基准：测语言模型在连续反馈下能不能越改越对"
title_en: "Seq@K Benchmark Tests Whether Language Models Improve Under Sequential Feedback"
url: https://ai.daily.yangsir.net/daily/20261101-T0-07
issue_date: 2026-11-01
publish_date: 2026-10-31T04:00:00.000Z
category: research
source_name: "arXiv cs.AI"
source_url: https://arxiv.org/abs/2610.37859
---

# Seq@K基准：测语言模型在连续反馈下能不能越改越对

arXiv新论文提出Seq@K基准，用于评估语言模型在顺序反馈下的表现。出发点是大模型代理区别于以往智能系统的能力——能从上下文中的自然语言学习，并据此反复修正答案，这是实际部署中常见的交互方式。Seq@K用K轮连续反馈来考察模型是否真的在改进，而不是只测单轮回答质量。对做AI代理产品的人来说，这类评测更贴近真实使用：用户会连续提意见，模型能不能收敛才是关键指标。

## English Version

**Seq@K Benchmark Tests Whether Language Models Improve Under Sequential Feedback**

A new arXiv paper introduces Seq@K, a benchmark for evaluating language models under sequential feedback. The premise is that LLM agents differ from prior systems in learning from natural language in context, enabling iterative refinement of answers, a common mode in real deployments. Seq@K measures whether models actually improve across K rounds of feedback rather than only scoring single-turn response quality. For teams building agent products, this style of evaluation tracks real usage more closely: users give feedback repeatedly, and whether the model converges is the key metric.

---

**来源**：[arXiv cs.AI](https://arxiv.org/abs/2610.37859)

**详情页**：https://ai.daily.yangsir.net/daily/20261101-T0-07

---

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