×
Home Current Archive Editorial board
Instructions for papers
For Authors Aim & Scope Contact
Original scientific article

CROSS DOMAIN LATENT SPACE ALIGNMENT FOR ZERO SHOT KNOWLEDGE TRANSFER IN LARGE SCALE NEURO SYMBOLIC FOUNDATION MODELS

By
Alijon Anvarov Orcid logo ,
Alijon Anvarov
Contact Alijon Anvarov

Associate Professor, Fergana Medical Institute of Public Health, Department of Uzbek and Foreign Languages , Fergana

Jonibek Berdikulov Orcid logo ,
Jonibek Berdikulov

Samarkand State Medical University , Samarkand , Uzbekistan

Azizjon Begaliev Orcid logo ,
Azizjon Begaliev

Department of Information Technology and Exact Sciences, Termez University of Economics and Service , Termez , Uzbekistan

Zulxumor Djurayeva Orcid logo ,
Zulxumor Djurayeva

Associate Professor, Bukhara State University , Bukhara , Uzbekistan

Khaydar Kamilov Orcid logo ,
Khaydar Kamilov

Tashkent State Medical University , Tashkent , Uzbekistan

Soniya Islyamova Orcid logo ,
Soniya Islyamova

Lecturer, Jizzakh State Pedagogical University , Jizzakh , Uzbekistan

Guzal Ergasheva Orcid logo
Guzal Ergasheva

Head, Department for Research, Innovation and Training of Scientific-Pedagogical Personnel, Alfraganus University , Tashkent , Uzbekistan

Abstract

The large-scale adoption of neuro-symbolic foundation models poses a critical challenge: zero-shot knowledge transfer is hindered by mismatches in latent distributions, modality heterogeneity, and discrepancies between symbolic and neural representations. In the following paper, a Cross-Domain Latent Space Alignment (CDLSA) framework will be proposed to support strong zero-shot generalisation across structurally different domains without task-specific fine-tuning. Its fundamental issue is the discrepancy between high-dimensional neural representations and structured symbolic representations that, under most circumstances, results in performance drops of 25-35 % when tested on unseen domains. The presented methodology combines contrastive latent projection, probabilistic regularisation of the manifold, and symbolic encoding of constraints into a single optimisation goal. The architecture uses a dual-encoder with shared latent anchors to reduce the Maximum Mean Discrepancy (MMD) and the Kullback-Leibler (KL) divergence between domains while maintaining semantic consistency. Multi-domain experimental assessments of vision-language and logic-reasoning have shown that  zero-shot accuracy increases by 21.8%, with a 17.3% error rate in cross-domain generalisation. Latent alignment minimised inter-domain distribution variance by 32.6% and had a higher symbolic consistency score of 0.84 (F1-measure) compared to the initial score of 0.68. Robustness (p<0.01) was tested across five independent runs, and the average effect size (Cohen's d) was 0.79, which is strong enough to be considered practical. Also, the calibration error was reduced by 14.5%, resulting in a more accurate estimate of uncertainty. The findings show that structured latent alignment provides a significant boost to knowledge transfer in neuro-symbolic systems in the zero-shot setting. The CDLSA framework is proposed to create a scalable pipeline from neural representation learning to symbolic reasoning, enabling more reliable and interpretable foundation models in heterogeneous data settings.

References

1.
Cunnington D, Law M, Lobo J, Russo A. The role of foundation models in neuro-symbolic learning and reasoning. InInternational Conference on Neural-Symbolic Learning and Reasoning 2024 Sep 9(pp. 84-100). Cham: Springer Nature Switzerland.
2.
Aho K, Roads BD, Love BC. System alignment supports cross-domain learning and zero-shot generalisation. Cognition. 2022 Oct 1; 227:105200.
3.
Ma H, Wang C, Song S, Yang S, Zhang L, Zhang X. Large language models are zero-shot cross-domain diagnosticians in cognitive diagnosis. Frontiers of Digital Education. 2025 Jun;2(2):17.
4.
Jha S, Roy A, Cobb A, Berenbeim A, Bastian ND. Challenges and opportunities in neuro-symbolic composition of foundation models. In MILCOM 2023-2023 IEEE Military Communications Conference (MILCOM) 2023 Oct 30 (pp. 156-161). IEEE.
5.
Xu S, Thomas CK, Hashash O, Muralidhar N, Saad W, Ramakrishnan N. Large multi-modal models (LMMs) as universal foundation models for AI-native wireless systems. IEEE Network. 2024 Jul 15;38(5):10-20.

Citation

This is an open access article distributed under the  Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. 

Article metrics

Google scholar: See link

The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.