Abstract:Graph learning aims to convert data into graph representations, which are fundamental to many problems in machine learning for CAD, where circuits, layouts, designs, and optimization states are often modeled as graph-structured objects. Existing graph learning methods usually rely on carefully designed graph construction rules, extensive parameter tuning, and sophisticated mathematical theory; moreover, achieving good performance often requires task-specific graph construction tailored to the downstream objective. In this work, we study whether a large language model (LLM) can reason about graph structure and infer a useful topology without observing the feature matrix, without knowing the downstream task, and without relying on any carefully designed graph construction algorithm or parameter tuning process. To this end, we propose T2T-LA, a Topology-to-Topology LLM Agent that receives no input other than a set of previously failed topologies and the scores assigned to them by a private scorer. The agent is not told what task or algorithm produces the scores, how these topologies are generated, or what the scores mean. Since none of the observed topologies is satisfactory, T2T-LA cannot simply imitate a good example. Instead, it is forced to infer hidden relationships between graph connectivity patterns and the observed scores, a capability that is particularly relevant to CAD scenarios where useful design structures may be difficult to specify manually. Experimental results show that T2T-LA can generate, in one shot, a graph topology that enables the downstream algorithm to produce a sufficiently good solution, suggesting a new LLM-driven direction for topology reasoning and graph representation learning in ML-for-CAD workflows.
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2512.08964 [cs.LG] |
| (or arXiv:2512.08964v4 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2512.08964 arXiv-issued DOI via DataCite |
From: Yongyu Wang [view email]
[v1]
Sun, 30 Nov 2025 01:40:15 UTC (22 KB)
[v2]
Wed, 17 Dec 2025 12:21:47 UTC (17 KB)
[v3]
Sat, 20 Dec 2025 10:29:58 UTC (414 KB)
[v4]
Fri, 15 May 2026 15:13:23 UTC (1,222 KB)
此内容由惯性聚合(RSS阅读器)自动聚合整理，仅供阅读参考。 原文来自 — 版权归原作者所有。