arXiv cs.CL· Vibhhu Sharma, Thorsten Joachims, Sarah Dean·· 12 小时前AI 评分42
Value Over Language Model（VOLM）：检测写作中的原创贡献
Value Over Language Model: Detecting Original Contribution in Writing
AI 导读
研究者提出 Value Over Language Model（VOLM）框架，无需训练或标注数据，通过在不同粒度上提取文档内容并用 LLM 重建，衡量人对文档的贡献相对于 LLM 仅凭任务描述即可产出内容的增量。在新闻文章、ICLR 同行评审和议论文三个领域，VOLM 能区分人类撰写文档与匹配的 LLM 生成文档，且对 LLM 重建、往返翻译等内容保持型变换基本不变。
正文 · 原文
Abstract:LLMs have been rapidly adopted across writing tasks, prompting the development of tools for detecting LLM-generated text. Yet, these tools largely measure how much of a document's surface text was written by an LLM and aren't fundamentally designed to measure how much of the information content or ideas originated from the LLM itself rather than being supplied by the user in the prompt. In this work, we design a framework that measures how much value a person adds on top of what a language model could have easily produced by itself. The method requires no training or labeled data and never scores the document's surface text, insulating it from stylistic confounders. Instead, it extracts the document's content at increasing levels of granularity, uses an LLM to reconstruct the document from each partial representation, and compares these reconstructions with those produced from the task description alone. We call this framework Value Over Language Model (VOLM), which measures a document's contribution relative to a replacement-level document that an LLM could produce from the task description alone. We evaluate VOLM with a specific instantiation of this framework across three domains: news articles, ICLR peer reviews, and argumentative essays. VOLM separates human-authored documents from matched LLM-generated documents produced from generic task descriptions, while remaining substantially invariant to content-preserving transformations, including LLM-based reconstruction and round-trip translation. We further find that increasingly constrained content extractors reduce residual differences between LLM-generated and humanized text, demonstrating the importance of disentangling informational content from stylistic variation. We hope these results encourage further work on specialized instantiations of the framework and on assessing human contributions in LLM-assisted writing more generally.
| Comments: | 38 pages |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2609.00700 [cs.AI] |
| (or arXiv:2609.00700v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2609.00700 arXiv-issued DOI via DataCite |
Submission history
From: Vibhhu Sharma [view email]
[v1]
Tue, 1 Sep 2026 04:21:20 UTC (5,563 KB)
[v2]
Wed, 4 Nov 2026 19:15:22 UTC (3,566 KB)
来源：arXiv cs.CL · arxiv.org