Curriculum Learning for Robust Reasoning in Small-Scale Transformer Models
Keywords:
curriculum learning; transformer models; reasoning robustness; small-scale models; deployment; fairness; sustainability; governanceAbstract
Large-scale transformer models exhibit strong reasoning capabilities, but their deployment in resource-constrained settings remains limited by computational cost, energy consumption, latency, and data governance requirements. Small-scale transformer models provide a tractable alternative, yet they frequently suffer from brittle reasoning, limited generalization, and sensitivity to spurious correlations. This paper examines curriculum learning as a systemic intervention for improving robust reasoning in compact transformer architectures. Rather than treating curriculum design as merely a scheduling heuristic, the analysis positions it within a broader socio-technical framework that includes architectural constraints, infrastructure trade-offs, robustness evaluation, fairness, alignment, deployment, and sustainability. The paper reviews evidence from foundational and recent work on progressive training schedules, reasoning path filtering, and alignment for small parameter language models. It argues that carefully staged exposure to concepts, tasks, and reasoning patterns can strengthen internal representations while reducing unnecessary computation. The discussion further addresses structural trade-offs involving memory footprint, throughput, distillation, and quantization, as well as governance and policy implications related to auditing, accountability, and equitable access. A central claim is that robust reasoning in small-scale systems is not primarily a matter of adding parameters or data; it requires coordinated design of training curricula, evaluation protocols, and deployment safeguards. The paper concludes by identifying open challenges and forward-looking research directions for building reasoning-capable small models that are more transparent, efficient, and aligned with institutional and societal expectations.
References
1. Bengio, Y., Louradour, J., Collobert, R., & Weston, J. (2009). Curriculum learning. In Proceedings of the 26th Annual International Conference on Machine Learning (pp. 41-48). ACM.
2. Elman, J. L. (1993). Learning and development in neural networks: The importance of starting small. Cognition, 48(1), 71-99.
3. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30.
4. Brown, T., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G., Henighan, T., Child, R., Ramesh, A., Ziegler, D., Wu, J., Winter, C., Hesse, C., Chen, M., Sigler, E., Litwin, M., Gray, S., Chess, B., Clark, J., Berner, C., McCandlish, S., Radford, A., Sutskever, I., & Amodei, D. (2020). Language models are few-shot learners. Advances in Neural Information Processing Systems, 33, 1877-1901.
5. Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019). DistilBERT, a distilled version of BERT: Smaller, faster, cheaper and lighter. arXiv preprint arXiv:1910.01108.
6. Kaplan, J., McCandlish, S., Henighan, T., Brown, T. B., Chess, B., Child, R., Gray, S., Radford, A., Wu, J., & Amodei, D. (2020). Scaling laws for neural language models. arXiv preprint arXiv:2001.08361.
7. Hoffmann, J., Borgeaud, S., Mensch, A., Buchatskaya, E., Cai, T., Rutherford, E., Casas, D. de L., Hendricks, L. A., Welbl, J., Clark, A., Hennigan, T., Noland, E., Millican, K., Driessche, G. van den, Damoc, B., Guy, A., Osindero, S., Simonyan, K., Elsen, E., Rae, J. W., Vinyals, O., & Sifre, L. (2022). Training compute-optimal large language models. arXiv preprint arXiv:2203.15556.
8. Bommasani, R., Hudson, D. A., Adeli, E., Altman, R., Arora, S., von Arx, S., Bernstein, M. S., Bohg, J., Bosselut, A., Brunskill, E., Brynjolfsson, E., Buch, S., Card, D., Castellon, R., Chatterji, N., Chen, A., Creel, K., Davis, J. Q., Demszky, D., Donahue, C., Doumbouya, M., Durmus, E., Ermon, S., Etchemendy, J., Ethayarajh, K., Fei-Fei, L., Finn, C., Gale, T., Gillespie, L., Goel, K., Goodman, N., Grossman, S., Guha, N., Hashimoto, T., Henderson, P., Hewitt, J., Ho, D. E., Hong, J., Hsu, K., Huang, J., Icard, T., Jain, S., Jurafsky, D., Kalluri, P., Karamcheti, S., Keeling, G., Khani, F., Khattab, O., Koh, P. W., Krass, M., Krishna,Kuditipudi, R., Kumar, A., Kusupati, A., Laidlaw, C., Lee, K., Li, D., Li, J., Li, X., Liang, P. (2021). On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258.
9. Li, Q. (2026, May). TrajLite PPO: Scalable Reasoning Path Filtering and Alignment for Small Parameter Large Language Models. In 2026 2nd International Conference on Artificial Intelligence and Digital Ethics (ICAIDE) (pp. 29-32). IEEE.
10. Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi, E., Le, Q., & Zhou, D. (2022). Chain-of-thought prompting elicits reasoning in large language models. Advances in Neural Information Processing Systems, 35, 24824-24837.
11. Nye, M., Andreassen, A. J., Gur-Ari, G., Michalewski, H., Austin, J., Bieber, D., Dohan, D., Lewkowycz, A., Bosma, M., Luan, D., Sutton, C., & Odena, A. (2021). Show your work: Scratchpads for intermediate computation with language models. arXiv preprint arXiv:2112.00114.
12. Rajpurkar, P., Jia, R., & Liang, P. (2018). Know what you don't know: Unanswerable questions for SQuAD. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) (pp. 784-789). Association for Computational Linguistics.
13. Bender, E. M., Gebru, T., McMillan-Major, A., & Shmitchell, S. (2021). On the dangers of stochastic parrots: Can language models be too big? In Proceedings of the 2021 ACM Conference on Fairness, Accountability, and Transparency (pp. 610-623). ACM.
14. Strubell, E., Ganesh, A., & McCallum, A. (2019). Energy and policy considerations for deep learning in NLP. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (pp. 3645-3650). Association for Computational Linguistics.
15. Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C. L., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., Schulman, J., Hilton, J., Kelton, F., Miller, L., Simens, M., Askell, A., Welinder, P., Christiano, P., Leike, J., & Lowe, R. (2022). Training language models to follow instructions with human feedback. Advances in Neural Information Processing Systems, 35, 27730-27744.
16. Hendrycks, D., Burns, C., Basart, S., Zou, A., Mazeika, M., Song, D., & Steinhardt, J. (2021). Measuring massive multitask language understanding. In International Conference on Learning Representations.
17. Zellers, R., Holtzman, A., Bisk, Y., Farhadi, A., & Choi, Y. (2019). HellaSwag: Can a machine really finish your sentence? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (pp. 4791-4800). Association for Computational Linguistics.
18. Cobbe, K., Kosaraju, V., Bavarian, M., Chen, M., Jun, H., Kaiser, L., Plappert, M., Tworek, J., Hilton, J., Nakano, R., Hesse, C., & Schulman, J. (2021). Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168.
19. Amodei, D., Olah, C., Steinhardt, J., Christiano, P., Schulman, J., & Mané, D. (2016). Concrete problems in AI safety. arXiv preprint arXiv:1606.06565.
20. Smith, L. N. (2017). Cyclical learning rates for training neural networks. In 2017 IEEE Winter Conference on Applications of Computer Vision (WACV) (pp. 464-472). IEEE.
21. Gururangan, S., Marasović, A., Swayamdipta, S., Lo, K., Beltagy, I., Downey, D., & Smith, N. A. (2020). Don't stop pretraining: Adapt language models to domains and tasks. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (pp. 8342-8360). Association for Computational Linguistics.
22. Wang, A., Singh, A., Michael, J., Hill, F., Levy, O., & Bowman, S. R. (2018). GLUE: A multi-task benchmark and analysis platform for natural language understanding. In International Conference on Learning Representations.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Journal of Engineering Systems and Digital Innovation
This work is licensed under a Creative Commons Attribution 4.0 International License.
