---
title: "人工修正：大语言模型为何过度使用一种经典修辞手法及其缓解方法"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2607.21498v1"
permalink: "https://feed.astralor.com/items/itm_ee89ef38d82f28630241aed5"
published: "2026-07-23T16:47:39.000Z"
score: 81
channel: "ai"
tags: ["LoRA适配器","修辞手法","大语言模型","强化学习人类反馈","文本风格校准"]
---

# 人工修正：大语言模型为何过度使用一种经典修辞手法及其缓解方法

## AI 点评

> 这篇论文让我眼前一亮——它把两千年前的修辞学与当代大模型训练巧妙结合，既学术又务实。对AI开发者来说，单行指令和LoRA适配器是现成的工具箱；对写作者和内容审核人员，它敲响了‘机器文风入侵’的警钟。别只被标题吸引，后面的风险警示值得每个依赖AI辅助写作的人认真思考。

## 速览

论文发现大语言模型系统性过度使用修辞手法epanorthosis（自我纠正），主要源于训练数据中的促销文本和RLHF偏好，生成模式是放大器。通过Epanorthosis Index测量，模型在演讲类文本中过度使用（约两倍，意大利语近三倍），非正式问答不足。提出缓解方法：单行指令可将该修辞减少一半至四分之三，监督微调适配器几乎完全消除，且可通过缩放系数调整到人类水平。目标是按体裁校准而非消除，真正风险是人类开始模仿机器写作。

## 关键点

- LLM系统性过度使用修辞手法epanorthosis（自我纠正），主要源于训练数据中的促销文本和RLHF偏好。
- 通过Epanorthosis Index测量，模型在演讲类文本中过度使用（约两倍，意大利语近三倍），而在非正式问答中不足。
- 缓解方法：单行指令可将该修辞减少一半至四分之三，监督微调适配器几乎完全消除，且可通过缩放系数调整到人类水平。
- 目标是按体裁校准到人类比率，而非完全消除；真正风险是人类开始模仿机器写作。

## 全文

A rhetorical figure that Cicero and Quintilian catalogued two thousand years ago reappears, systematically, in the text of large language models: epanorthosis, the self-correction of the specimen «This is not a course. It is a journey of transformation». This essay argues that the overuse is a trained disposition, driven mainly by a training distribution rich in promotional prose and by preference tuning (RLHF) that rewards confident, emphatic phrasing; the left-to-right nature of generation is an amplifier rather than the root cause. Building on evidence that models diverge from human rhetorical style, and on Fontanier's classification of epanorthosis as a figure of thought, it sets out a programme that scores the figure against genre-specific human baselines through an Epanorthosis Index (density relative to the human rate). A first measurement, on three sizes of one instruction-tuned model family, finds mis-calibration by register in both directions: the models overshoot in oratory (about twofold, near threefold in Italian, concentrated in the larger tiers) and undershoot in informal question-and-answer writing, while matching humans in argument, journalism, and encyclopedic prose. Three constructive contributions follow: a survey of mitigation techniques centred on lightweight LoRA adapters; a demonstration, in Italian, that a one-line instruction cuts the figure by half to nearly three-quarters and that a supervised-fine-tuning adapter removes it almost entirely, with a scaling coefficient that dials the reduction back onto the human rate; and the argument that the target is calibration to the human rate for each genre, not elimination. It closes on the stakes: the real risk is that we begin to write like the machines.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_ee89ef38d82f28630241aed5