大模型
arxiv.org ·
非你所指：LLM 能否遵循指定的否定语义？
论文研究大语言模型默认采用哪种否定语义解读，以及能否在明确指定时覆盖该偏好，并为此提出 NAFBench——一个程序化生成求解器认证实例的基准，覆盖 SLDNF、良基语义（WFS）以及稳定模型语义下的轻信与怀疑推理四种视角。生成器产出受控深度、宽度和环结构的基正常逻辑程序，用 SWI-Prolog、良基语义求解器和 clingo 在四种视角下求解，得到最多四种分歧标签，再以多种框架和规则顺序自然语言化。结果显示存在一致差距：开源模型遵循指定否定语义仍未解决，最强模型在四种语义视角下得分 59–74%，最弱 31–67%；所有模型在超过一半逻辑等价的规则重排上顺序敏感，两个较弱模型常在良基“未定义”上过度断言。两个前沿模型在主固定复杂度评测集上达 100%，o4-mini 接近满分，仅在良基“未定义”上降至 81%。将推理委托给求解器、在认证轨迹上微调或强制显式三值判定均可部分缩小差距。
AI推荐理由 arXiv 新论文（2026-09-23 提交），提出 NAFBench 程序化生成器，用 SWI-Prolog、WFS 求解器和 clingo 生成求解器认证的实例，覆盖 SLDNF、良基语义（WFS）以及稳定模型语义下的轻信/怀疑推理四种否定语义视角，并给出可验证的量化结果：最强开源模型在四种语义视角下得分 59–74%，最弱 31–67%，所有模型在超过一半逻辑等价的规则重排上表现出顺序敏感性，两个较弱模型常在良基“未定义”上过度断言；两个前沿模型在主固定复杂度评测集上达 100%，o4-mini 接近满分但在良基“未定义”上降至 81%；将推理委托给求解器、在认证轨迹上微调或强制显式三值判定可部分缩小差距。局限是仅摘要级信息，未提供数据集规模、模型清单与统计细节，且评测集为固定复杂度，泛化性待验证。
78
正文
Computer Science > Artificial Intelligence
arXiv:2609.27517 (cs)
[Submitted on 23 Sep 2026]
Title:Not What You Meant: Can LLMs Follow a Specified Negation Semantics?
Authors:Qiming Bao, Agnieszka Mensfelt, Michael J. Witbrock, Kostas Stathis
View a PDF of the paper titled Not What You Meant: Can LLMs Follow a Specified Negation Semantics?, by Qiming Bao and 3 other authors
View PDF
HTML (experimental)
Abstract:Negation does not carry a uniform interpretation across domains. In legal, regulatory, and medical reasoning, the intended interpretation depends on the reading in force -- open- versus closed-world, two- versus three-valued, and credulous versus skeptical. We study which reading of negation large language models adopt by default and whether they can override that preference when a different reading is explicitly specified. To this end, we introduce NAFBench, a procedural generator of solver-certified instances spanning four semantic viewpoints: SLDNF, well-founded semantics (WFS), and credulous and skeptical reasoning under stable-model semantics. The generator emits ground normal logic programs with controlled depth, width, and cycle structure. Each program is solved under all four viewpoints using SWI-Prolog, a well-founded semantics solver, and clingo, yielding up to four divergent labels. The programs are then verbalized into natural language under multiple framings and rule orderings that leave the answer invariant. The results expose a consistent gap. Across open-source models, following a specified negation semantics remains unsolved: the strongest models score 59--74% across the four semantic viewpoints, while the weakest score 31--67%. All models are order-sensitive on more than half of logically identical rule shufflings, while the two weaker models frequently overcommit on well-founded "undefined." Two frontier models reach 100% on the main fixed-complexity evaluation set, and a third, o4-mini, is near-perfect, falling only to 81% on well-founded "undefined." Delegating reasoning to a solver, fine-tuning on certified traces, or forcing an explicit three-valued verdict each partly closes the gap.
Subjects:
Artificial Intelligence (cs.AI)
Cite as:
arXiv:2609.27517 [cs.AI]
(or
arXiv:2609.27517v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2609.27517
Focus to learn more
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Qiming Bao [view email]
[v1]
Wed, 23 Sep 2026 08:14:47 UTC (48 KB)
Full-text links:
Access Paper:
View a PDF of the paper titled Not What You Meant: Can LLMs Follow a Specified Negation Semantics?, by Qiming Bao and 3 other authors
View PDF
HTML (experimental)
TeX Source
view license
Current browse context:
cs.AI
new
|
recent
| 2026-09
Change to browse by:
cs
References & Citations
NASA ADS
Google Scholar
Semantic Scholar
export BibTeX citation
Loading...
BibTeX formatted citation
×
loading...
Data provided by:
Bookmark
Bibliographic Tools
Bibliographic and Citation Tools
Bibliographic Explorer Toggle
Bibliographic Explorer (What is the Explorer?)
Connected Papers Toggle
Connected Papers (What is Connected Papers?)
Litmaps Toggle
Litmaps (What is Litmaps?)
scite.ai Toggle
scite Smart Citations (What are Smart Citations?)
Code, Data, Media
Code, Data and Media Associated with this Article
alphaXiv Toggle
alphaXiv (What is alphaXiv?)
Links to Code Toggle
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub Toggle
DagsHub (What is DagsHub?)
GotitPub Toggle
Gotit.pub (What is GotitPub?)
Huggingface Toggle
Hugging Face (What is Huggingface?)
ScienceCast Toggle
ScienceCast (What is ScienceCast?)
Demos
Demos
Replicate Toggle
Replicate (What is Replicate?)
Spaces Toggle
Hugging Face Spaces (What is Spaces?)
Spaces Toggle
TXYZ.AI (What is TXYZ.AI?)
Related Papers
Recommenders and Search Tools
Link to Influence Flower
Influence Flower (What are Influence Flowers?)
Core recommender toggle
CORE Recommender (What is CORE?)
Author
Venue
Institution
Topic
About arXivLabs
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Which authors of this paper are endorsers? |
Disable MathJax (What is MathJax?)