Paper
CayleyPy-4: AI-Holography. Towards analogs of holographic string dualities for AI tasks
Authors
A. Chervov, F. Levkovich-Maslyuk, A. Smolensky, F. Khafizov, I. Kiselev, D. Melnikov, I. Koltsov, S. Kudashev, D. Shiltsov, M. Obozov, S. Krymskii, V. Kirova, E. V. Konstantinova, A. Soibelman, S. Galkin, L. Grunwald, A. Kotov, A. Alexandrov, S. Lytkin, D. Fedoriaka, A. Chevychelov, Z. Kogan, A. Natyrova, L. Cheldieva, O. Nikitina, S. Fironov, A. Vakhrushev, A. Lukyanenko, V. Ilin, D. Gorodkov, N. Bogachev, I. Gaiur, M. Zaitsev, F. Petrov, L. Petrov, T. Gaintseva, A. Gavrilova, M. N. Smirnov, N. Kalinin, A. Khan, K. Jung, H. Mousset, H. Isambert, O. Debeaupuis
Abstract
This is the fourth paper in the CayleyPy project, which applies AI methods to the exploration of large graphs. In this work, we suggest the existence of a new discrete version of holographic string dualities for this setup, and discuss their relevance to AI systems and mathematics. Many modern AI tasks -- such as those addressed by GPT-style language models or RL systems -- can be viewed as direct analogues of predicting particle trajectories on graphs. We investigate this problem for a large family of Cayley graphs, for which we show that surprisingly it admits a dual description in terms of discrete strings. We hypothesize that such dualities may extend to a range of AI systems where they can lead to more efficient computational approaches. In particular, string holographic images of states are proposed as natural candidates for data embeddings, motivated by the "complexity = volume" principle in AdS/CFT. For Cayley graphs of the symmetric group S_n, our results indicate that the corresponding dual objects are flat, planar polygons. The diameter of the graph is equal to the number of integer points inside the polygon scaled by n. Vertices of the graph can be mapped holographically to paths inside the polygon, and the usual graph distances correspond to the area under the paths, thus directly realising the "complexity = volume" paradigm. We also find evidence for continuous CFTs and dual strings in the large n limit. We confirm this picture and other aspects of the duality in a large initial set of examples. We also present new datasets (obtained by a combination of ML and conventional tools) which should be instrumental in establishing the duality for more general cases.
Metadata
Related papers
Vibe Coding XR: Accelerating AI + XR Prototyping with XR Blocks and Gemini
Ruofei Du, Benjamin Hersh, David Li, Nels Numan, Xun Qian, Yanhe Chen, Zhongy... • 2026-03-25
Comparing Developer and LLM Biases in Code Evaluation
Aditya Mittal, Ryan Shar, Zichu Wu, Shyam Agarwal, Tongshuang Wu, Chris Donah... • 2026-03-25
The Stochastic Gap: A Markovian Framework for Pre-Deployment Reliability and Oversight-Cost Auditing in Agentic Artificial Intelligence
Biplab Pal, Santanu Bhattacharya • 2026-03-25
Retrieval Improvements Do Not Guarantee Better Answers: A Study of RAG for AI Policy QA
Saahil Mathur, Ryan David Rittner, Vedant Ajit Thakur, Daniel Stuart Schiff, ... • 2026-03-25
MARCH: Multi-Agent Reinforced Self-Check for LLM Hallucination
Zhuo Li, Yupeng Zhang, Pengyu Cheng, Jiajun Song, Mengyu Zhou, Hao Li, Shujie... • 2026-03-25
Raw Data (Debug)
{
"raw_xml": "<entry>\n <id>http://arxiv.org/abs/2603.22195v1</id>\n <title>CayleyPy-4: AI-Holography. Towards analogs of holographic string dualities for AI tasks</title>\n <updated>2026-03-23T16:54:44Z</updated>\n <link href='https://arxiv.org/abs/2603.22195v1' rel='alternate' type='text/html'/>\n <link href='https://arxiv.org/pdf/2603.22195v1' rel='related' title='pdf' type='application/pdf'/>\n <summary>This is the fourth paper in the CayleyPy project, which applies AI methods to the exploration of large graphs. In this work, we suggest the existence of a new discrete version of holographic string dualities for this setup, and discuss their relevance to AI systems and mathematics. Many modern AI tasks -- such as those addressed by GPT-style language models or RL systems -- can be viewed as direct analogues of predicting particle trajectories on graphs. We investigate this problem for a large family of Cayley graphs, for which we show that surprisingly it admits a dual description in terms of discrete strings. We hypothesize that such dualities may extend to a range of AI systems where they can lead to more efficient computational approaches. In particular, string holographic images of states are proposed as natural candidates for data embeddings, motivated by the \"complexity = volume\" principle in AdS/CFT.\n For Cayley graphs of the symmetric group S_n, our results indicate that the corresponding dual objects are flat, planar polygons. The diameter of the graph is equal to the number of integer points inside the polygon scaled by n. Vertices of the graph can be mapped holographically to paths inside the polygon, and the usual graph distances correspond to the area under the paths, thus directly realising the \"complexity = volume\" paradigm. We also find evidence for continuous CFTs and dual strings in the large n limit. We confirm this picture and other aspects of the duality in a large initial set of examples. We also present new datasets (obtained by a combination of ML and conventional tools) which should be instrumental in establishing the duality for more general cases.</summary>\n <category scheme='http://arxiv.org/schemas/atom' term='hep-th'/>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n <category scheme='http://arxiv.org/schemas/atom' term='math.CO'/>\n <category scheme='http://arxiv.org/schemas/atom' term='math.GR'/>\n <published>2026-03-23T16:54:44Z</published>\n <arxiv:comment>20+120 pages</arxiv:comment>\n <arxiv:primary_category term='hep-th'/>\n <author>\n <name>A. Chervov</name>\n </author>\n <author>\n <name>F. Levkovich-Maslyuk</name>\n </author>\n <author>\n <name>A. Smolensky</name>\n </author>\n <author>\n <name>F. Khafizov</name>\n </author>\n <author>\n <name>I. Kiselev</name>\n </author>\n <author>\n <name>D. Melnikov</name>\n </author>\n <author>\n <name>I. Koltsov</name>\n </author>\n <author>\n <name>S. Kudashev</name>\n </author>\n <author>\n <name>D. Shiltsov</name>\n </author>\n <author>\n <name>M. Obozov</name>\n </author>\n <author>\n <name>S. Krymskii</name>\n </author>\n <author>\n <name>V. Kirova</name>\n </author>\n <author>\n <name>E. V. Konstantinova</name>\n </author>\n <author>\n <name>A. Soibelman</name>\n </author>\n <author>\n <name>S. Galkin</name>\n </author>\n <author>\n <name>L. Grunwald</name>\n </author>\n <author>\n <name>A. Kotov</name>\n </author>\n <author>\n <name>A. Alexandrov</name>\n </author>\n <author>\n <name>S. Lytkin</name>\n </author>\n <author>\n <name>D. Fedoriaka</name>\n </author>\n <author>\n <name>A. Chevychelov</name>\n </author>\n <author>\n <name>Z. Kogan</name>\n </author>\n <author>\n <name>A. Natyrova</name>\n </author>\n <author>\n <name>L. Cheldieva</name>\n </author>\n <author>\n <name>O. Nikitina</name>\n </author>\n <author>\n <name>S. Fironov</name>\n </author>\n <author>\n <name>A. Vakhrushev</name>\n </author>\n <author>\n <name>A. Lukyanenko</name>\n </author>\n <author>\n <name>V. Ilin</name>\n </author>\n <author>\n <name>D. Gorodkov</name>\n </author>\n <author>\n <name>N. Bogachev</name>\n </author>\n <author>\n <name>I. Gaiur</name>\n </author>\n <author>\n <name>M. Zaitsev</name>\n </author>\n <author>\n <name>F. Petrov</name>\n </author>\n <author>\n <name>L. Petrov</name>\n </author>\n <author>\n <name>T. Gaintseva</name>\n </author>\n <author>\n <name>A. Gavrilova</name>\n </author>\n <author>\n <name>M. N. Smirnov</name>\n </author>\n <author>\n <name>N. Kalinin</name>\n </author>\n <author>\n <name>A. Khan</name>\n </author>\n <author>\n <name>K. Jung</name>\n </author>\n <author>\n <name>H. Mousset</name>\n </author>\n <author>\n <name>H. Isambert</name>\n </author>\n <author>\n <name>O. Debeaupuis</name>\n </author>\n </entry>"
}