---
title: "TeamOmega at SemEval-2026 Task 13: Frozen vs. Trainable"
description: "The difficulty comes from the test set. 0.9821 recall on AI. 21 references. Among them: 0."
authors: ["Nahid Niyaz Shovon", "Md. Naim Parvez"]
year: 2026
venue: "Proceedings of the 20th International Workshop on Semantic Evaluation (2026)"
source: "https://aclanthology.org/2026.semeval-1.255/"
doi: "10.18653/v1/2026.semeval-1.255"
licence: "CC BY 4.0 (ACL Anthology)"
licence_url: "https://creativecommons.org/licenses/by/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/shovon-2026-teamomega-semeval-2026-task-13/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/shovon-2026-teamomega-semeval-2026-task-13/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/52/) / TeamOmega at SemEval-2026 Task 13: Frozen vs. Trainable Representations for Out-of-Distribution AI-Generated Code Detection: A CodeBERT Fine-Tuning Study

# TeamOmega at SemEval-2026 Task 13: Frozen vs. Trainable Representations for Out-of-Distribution AI-Generated Code Detection: A CodeBERT Fine-Tuning Study

**Authors:** Nahid Niyaz Shovon and Md. Naim Parvez · **Year:** 2026 · **Venue:** [Proceedings of the 20th International Workshop on Semantic Evaluation (2026)](https://md.research.elamur.ai/free/en/venues/semeval/) ([SemEval 2026](https://md.research.elamur.ai/free/en/venues/semeval/2026/)) · **Publisher:** [Association for Computational Linguistics](https://md.research.elamur.ai/free/en/publishers/association-for-computational-linguistics/) · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.semeval-1.255/) · [PDF](https://aclanthology.org/2026.semeval-1.255.pdf) · [DOI 10.18653/v1/2026.semeval-1.255](https://doi.org/10.18653/v1/2026.semeval-1.255) · [OpenAlex](https://openalex.org/W7166883377)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
The paper reports on Team_Omega's entry for Subtask A of SemEval-2026 Task 13. The subtask is a binary decision: was a code snippet written by a human or produced by a machine? The difficulty comes from the test set. Training data covers only algorithmic problems in C++, Python and Java. The hidden test set goes further, bringing in programming languages absent from training (Go, PHP, C#, C, JavaScript) along with code from domains not seen in training, namely research and production code.

The authors compare two CodeBERT setups:
- **Frozen encoder:** the whole encoder stays fixed, and only a small MLP head on the [CLS] vector is trained.
- **Partial unfreezing:** the final encoder layer is trained with a lower learning rate, under a deeper residual head that uses GELU, LayerNorm and class-weighted loss.

Each model saw the 500K training samples once. Adapting the encoder raised in-domain validation macro F1 considerably, but on the out-of-distribution leaderboard it did much worse. There it collapsed toward labelling nearly everything as AI-generated. The frozen model scored lower in-domain yet held up better under the shift.

The takeaway is that strong in-domain scores can be misleading for code-origin detection. Keeping pretrained representations intact may transfer better across languages than aggressive task adaptation.

## Key points
- **Data:** 500,000 training samples (238,475 human, 261,525 AI), about 100K for validation, and a 500K hidden test set. A 1,000-example labeled sample was used only for sanity checks.
- **Configuration A:** frozen microsoft/codebert-base with a head of about 214K parameters (768→256→64→2, ReLU).
- **Configuration B:** layer 11 unfrozen, with about 8.2M trainable parameters. The encoder learning rate was 2×10⁻⁵ and the head's was 2×10⁻⁴.
- **Results (validation → test macro F1):** A went from 0.9216 to 0.5132, while B went from 0.9841 to 0.3026. The team placed 44th of 81 in Subtask A.
- **Error analysis on the 1,000-sample check:** B reached 0.9821 recall on AI code but only 0.1287 on human code.
- **Caveats noted by the authors:** B may have been undertrained after a single epoch, a limit set by Kaggle's 12-hour GPU budget on a P100. Class weighting is a confounding difference between the setups. They suggest contrastive learning or adversarial domain adaptation as next steps.

*Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.*

## Institutions

- [Rajshahi University of Engineering and Technology](https://md.research.elamur.ai/free/en/universities/rajshahi-university-of-engineering-and-technology/) — University, Rajshahi, Bangladesh · 3 works

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 17,965 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 17,450 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 17,296 works · [🗄️ Information Systems](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · 1,730 works

## Cite

Cite as: Nahid Niyaz Shovon and Md. Naim Parvez. 2026. TeamOmega at SemEval-2026 Task 13: Frozen vs. Trainable Representations for Out-of-Distribution AI-Generated Code Detection: A CodeBERT Fine-Tuning Study. Proceedings of the 20th International Workshop on Semantic Evaluation (2026).

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{shovon2026teamomega,
  title = {{TeamOmega at SemEval-2026 Task 13: Frozen vs. Trainable Representations for Out-of-Distribution AI-Generated Code Detection: A CodeBERT Fine-Tuning Study}},
  author = {Nahid Niyaz Shovon and Md. Naim Parvez},
  booktitle = {Proceedings of the 20th International Workshop on Semantic Evaluation (2026)},
  year = {2026},
  doi = {10.18653/v1/2026.semeval-1.255},
  url = {https://doi.org/10.18653/v1/2026.semeval-1.255}
}
```

- [← previous TeamFusion: Supporting Open-ended Teamwork with Multi-Agent Systems](https://md.research.elamur.ai/free/en/papers/liu-2026-teamfusion-supporting-open-ended-teamwork/)
- [next → TeleAI at SemEval-2026 Task 13: Data-Centric Full-Parameter Fine-Tuning with Multi-Level Ensembling for Generated Code Detection](https://md.research.elamur.ai/free/en/papers/wang-2026-teleai-semeval-2026-task-13/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/52/)

## References

21 references, 7 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Wasi U Ahmad, Snigdha Chakraborty, and 1 others. 2024. Out-of-distribution generalization for code authorship attribution. In Findings of the Association for Computational Linguistics: EMNLP 2024. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Wasi+U+Ahmad%2C+Snigdha+Chakraborty%2C+and+1+others%2E+2024%2E+Out%2Dof%2Ddistribution+generalization+for+code+authorship+attribution%2E+In+Findings+of+the+Association+for+Computational+Linguistics%3A+EMNLP+2024%2E+Ass)
2. Jimmy Lei Ba, Jamie Ryan Kiros, and Geoffrey E. Hinton. 2016. [Layer normalization](https://md.research.elamur.ai/free/en/papers/ba-2016-layer-normalization/). arXiv preprint arXiv:1607.06450.
3. Mark Chen, Jerry Tworek, Heewoon Jun, Qiming Yuan, Henrique Peneder Pinto, Jared Kaplan, Harri Edwards, Yura Burda, Nicholas Joseph, Greg Brockman, and 1 others. 2021. [Evaluating large language models trained on code](https://md.research.elamur.ai/free/en/papers/chen-2021-evaluating-large-language-models-trained/). arXiv preprint arXiv:2107.03374.
4. Zhangyin Feng, Daya Guo, Duyu Tang, Nan Duan, Xiaocheng Feng, Ming Gong, Linjun Shou, Bing Qin, Ting Liu, Daxin Jiang, and Ming Zhou. 2020. Code-BERT: A pre-trained model for programming and natural languages. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 1536–1547. [find](https://scholar.google.com/scholar?q=Zhangyin+Feng%2C+Daya+Guo%2C+Duyu+Tang%2C+Nan+Duan%2C+Xiaocheng+Feng%2C+Ming+Gong%2C+Linjun+Shou%2C+Bing+Qin%2C+Ting+Liu%2C+Daxin+Jiang%2C+and+Ming+Zhou%2E+2020%2E+Code%2DBERT%3A+A+pre%2Dtrained+model+for+programming+and+natural+l)
5. Xavier Glorot and Yoshua Bengio. 2010. Understanding the difficulty of training deep feedforward neural networks. In Proceedings of the 13th International Conference on Artificial Intelligence and Statistics. [find](https://scholar.google.com/scholar?q=Xavier+Glorot+and+Yoshua+Bengio%2E+2010%2E+Understanding+the+difficulty+of+training+deep+feedforward+neural+networks%2E+In+Proceedings+of+the+13th+International+Conference+on+Artificial+Intelligence+and+Sta)
6. Daya Guo, Shuo Ren, Shuai Lu, Shengdong Li, Duyu Feng, Dazhao Liu, Ming Li, Ming Zhou, Nan Duan, and Alex Zhou. 2022. [UniXcoder: Unified cross-modal pre-training for code representation](https://md.research.elamur.ai/free/en/papers/guo-2022-unixcoder-unified-cross-modal-pre/). In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers…
7. Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. 2016. [Deep residual learning for image recognition](https://md.research.elamur.ai/free/en/papers/he-2015-deep-residual-learning-image-recognition/). In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition.
8. Dan Hendrycks and Kevin Gimpel. 2016. Gaussian error linear units (gelus). arXiv preprint arXiv:1606.08415. [find](https://scholar.google.com/scholar?q=Dan+Hendrycks+and+Kevin+Gimpel%2E+2016%2E+Gaussian+error+linear+units+%28gelus%29%2E+arXiv+preprint+arXiv%3A1606%2E08415%2E)
9. James Kirkpatrick, Razvan Pascanu, Neil Rabinowitz, Joel Veness, Guillaume Desjardins, Andrei A Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, and 1 others. 2017. Overcoming catastrophic forgetting in neural networks. Proceedings of the National Academy of Sciences. [find](https://scholar.google.com/scholar?q=James+Kirkpatrick%2C+Razvan+Pascanu%2C+Neil+Rabinowitz%2C+Joel+Veness%2C+Guillaume+Desjardins%2C+Andrei+A+Rusu%2C+Kieran+Milan%2C+John+Quan%2C+Tiago+Ramalho%2C+Agnieszka+Grabska%2DBarwinska%2C+and+1+others%2E+2017%2E+Overcomin)
10. Clara Kübler, Pasquale Minervini, and Sebastian Riedel. 2024. Semeval-2024 task 1: Track A (human vs. AI text detection). In Proceedings of the 18th International Workshop on Semantic Evaluation (SemEval-2024). Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Clara+K%C3%BCbler%2C+Pasquale+Minervini%2C+and+Sebastian+Riedel%2E+2024%2E+Semeval%2D2024+task+1%3A+Track+A+%28human+vs%2E+AI+text+detection%29%2E+In+Proceedings+of+the+18th+International+Workshop+on+Semantic+Evaluation+%28SemE)
11. Ilya Loshchilov and Frank Hutter. 2019. Decoupled weight decay regularization. In International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Ilya+Loshchilov+and+Frank+Hutter%2E+2019%2E+Decoupled+weight+decay+regularization%2E+In+International+Conference+on+Learning+Representations%2E)
12. Eric Mitchell, Jingfeng Peng, Timo Dettmers, Daniel Krasnopolsky, Elaine A. Chi, Christopher Potts, and Eric Horvitz. 2023. [DetectGPT: Zero-shot machine-generated text detection using probability curvature](https://md.research.elamur.ai/free/en/papers/mitchell-2023-detectgpt-zero-shot-machine-generated/). Transactions on Machine Learning Research.
13. Erik Nijkamp, Bo Li, Mark Rabe, and 1 others. 2023. Codegen: Open large language models for code. In International Conference on Learning Representations. [find](https://scholar.google.com/scholar?q=Erik+Nijkamp%2C+Bo+Li%2C+Mark+Rabe%2C+and+1+others%2E+2023%2E+Codegen%3A+Open+large+language+models+for+code%2E+In+International+Conference+on+Learning+Representations%2E)
14. OpenAI. 2023. OpenAI classifier. [https://openai.com/blog/new-ai-classifier-for-indicatin](https://openai.com/blog/new-ai-classifier-for-indicatin) g-ai-written-text. Accessed: 2026-03-01. [↗](https://openai.com/blog/new-ai-classifier-for-indicatin)
15. Daniil Orel, Dilshod Azizov, and Preslav Nakov. 2025a. CoDet-M4: Detecting machine-generated code in multi-lingual, multi-generator and multi-domain settings. In Findings of the Association for Computational Linguistics: ACL 2025, pages 10570–10593. [find](https://scholar.google.com/scholar?q=Daniil+Orel%2C+Dilshod+Azizov%2C+and+Preslav+Nakov%2E+2025a%2E+CoDet%2DM4%3A+Detecting+machine%2Dgenerated+code+in+multi%2Dlingual%2C+multi%2Dgenerator+and+multi%2Ddomain+settings%2E+In+Findings+of+the+Association+for+Comput)
16. Daniil Orel, Dilshod Azizov, Indraneil Paul, Yuxia Wang, Iryna Gurevych, and Preslav Nakov. 2026. [SemEval-2026 task 13: Detecting machine-generated code with multiple programming languages, generators, and application scenarios](https://md.research.elamur.ai/free/en/papers/orel-2026-semeval-2026-task-13-detecting/). In Proceedings of the 20th International Workshop on Semantic Evaluatio…
17. Daniil Orel, Indraneil Paul, Iryna Gurevych, and Preslav Nakov. 2025b. Droid: A resource suite for AI-generated code detection. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 31263–31289. [find](https://scholar.google.com/scholar?q=Daniil+Orel%2C+Indraneil+Paul%2C+Iryna+Gurevych%2C+and+Preslav+Nakov%2E+2025b%2E+Droid%3A+A+resource+suite+for+AI%2Dgenerated+code+detection%2E+In+Proceedings+of+the+2025+Conference+on+Empirical+Methods+in+Natural+La)
18. Shuo Ren, Daya Guo, Shuai Lu, Shengdong Li, Duyu Feng, Dazhao Liu, Ming Li, Ming Zhou, Nan Duan, and Alex Zhou. 2023. Cross-lingual code representation learning. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). Association for Comput… [find](https://scholar.google.com/scholar?q=Shuo+Ren%2C+Daya+Guo%2C+Shuai+Lu%2C+Shengdong+Li%2C+Duyu+Feng%2C+Dazhao+Liu%2C+Ming+Li%2C+Ming+Zhou%2C+Nan+Duan%2C+and+Alex+Zhou%2E+2023%2E+Cross%2Dlingual+code+representation+learning%2E+In+Proceedings+of+the+61st+Annual+Meet)
19. Leslie N. Smith. 2018. A disciplined approach to neural network hyper-parameters: Part 1 – learning rate, batch size, momentum, and weight decay. arXiv preprint arXiv:1803.09820. — [arXiv:1803.09820](https://arxiv.org/abs/1803.09820) [find](https://scholar.google.com/scholar?q=Leslie+N%2E+Smith%2E+2018%2E+A+disciplined+approach+to+neural+network+hyper%2Dparameters%3A+Part+1+%E2%80%93+learning+rate%2C+batch+size%2C+momentum%2C+and+weight+decay%2E+arXiv+preprint+arXiv%3A1803%2E09820%2E+%E2%80%94+%5BarXiv%3A1803%2E09820%5D%28)
20. Yue Wang, Weishi Li, Marek Mishra, Shomir Goldwasser, Ding Wang, and Fei Xia. 2021. CodeT5: Identifier-aware unified pre-trained encoder-decoder models for code understanding and generation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th Inter… [find](https://scholar.google.com/scholar?q=Yue+Wang%2C+Weishi+Li%2C+Marek+Mishra%2C+Shomir+Goldwasser%2C+Ding+Wang%2C+and+Fei+Xia%2E+2021%2E+CodeT5%3A+Identifier%2Daware+unified+pre%2Dtrained+encoder%2Ddecoder+models+for+code+understanding+and+generation%2E+In+Procee)
21. Qinkai Zheng, Xiao Xia, Xu Zou, Yuxiao Dong, Shan Wang, Yufei Xue, Zihan Wang, Lei Shen, Andi Wang, Yang Li, Teng Su, Zhilin Yang, and Jie Tang. 2023. [Codegeex: A pre-trained model for code generation with multilingual benchmarking on humaneval-x](https://md.research.elamur.ai/free/en/papers/zheng-2023-codegeex-pre-trained-model-code/). arXiv preprint arXiv:2303.17568.

*Updated 1 November 2026*