arXiv：cs.LG· Stefan Szeider·· 4 小时前AI 评分34
ZeroFolio：用文本嵌入实现零领域知识的算法选择
Algorithm Selection with Zero Domain Knowledge via Text Embeddings
AI 导读
研究者提出 ZeroFolio，一种无需手工特征的算法选择方法，直接将原始实例文件作为纯文本输入预训练文本嵌入模型，再通过加权 k 近邻选择算法。在覆盖 SAT、MaxSAT、QBF、ASP、CSP、MIP 和图问题共 7 个领域的 11 个 ASlib 场景中，ZeroFolio 在 9 个场景上优于基于手工特征的随机森林；与逐场景调优的随机森林相比则赢下 8 个场景。
正文
Abstract:We propose ZeroFolio, a feature-free approach to algorithm selection that uses pretrained text embeddings instead of hand-crafted instance features. It reads the raw instance file as plain text, embeds it with a pretrained embedding model, and selects an algorithm via weighted k-nearest neighbors. Our approach is based on the observation that pretrained embeddings can distinguish problem instances without any domain knowledge or task-specific training. ZeroFolio applies to any problem domain with text-based instance formats. We evaluate our approach on 11 ASlib scenarios spanning 7 domains (SAT, MaxSAT, QBF, ASP, CSP, MIP, and graph problems). ZeroFolio outperforms a random forest trained on hand-crafted features in 9 of 11 scenarios, often substantially, and in 8 of them with every serialization seed. It wins 8 of 11 scenarios against a per-scenario-tuned random forest. On the three scenarios with published AutoFolio results from the 2015 ICON Challenge, ZeroFolio comes within a small margin of AutoFolio without any per-scenario tuning. Our ablation study on SAT12-ALL shows that inverse-distance weighting and line shuffling improve performance. We further analyze the sensitivity of our approach to the serialization seed. On the SAT12-ALL scenario, where the random forest is stronger, both methods can be combined via soft voting to achieve further improvements.
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Machine Learning (cs.LG) |
| Cite as: | arXiv:2604.19753 [cs.AI] |
| (or arXiv:2604.19753v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2604.19753 arXiv-issued DOI via DataCite |
Submission history
From: Stefan Szeider [view email]
[v1]
Fri, 20 Mar 2026 13:07:59 UTC (299 KB)
[v2]
Sat, 11 Jul 2026 09:49:29 UTC (307 KB)
[v3]
Mon, 5 Oct 2026 19:26:23 UTC (307 KB)
来源：arXiv：cs.LG · arxiv.org