Boundary-Focused Large Language Model Adaptation for Style Change Detection in Multi-Authored Text
Source abstract
Abstract
The style change detection (SCD) task involves identifying the locations of writing style changes in multi-authored documents. This task can be applied to plagiarism detection, security, and commerce applications. Introducing decoder-based Large Language Models (LLMs) marks a pivotal shift in applications. The segment boundaries for SCD models can be represented by concatenating two consecutive segments as pairs. However, LLMs usually restrict their input lengths, where the long-length inputs may exceed the restricted length. This paper seeks to bridge this gap and exploit the power of LLMs by introducing boundary-focused LLM Adaptation for SCD (BF-LLMA-SCD). The proposed solution adapts decoder-based LLMs for SCD using QLoRA. BF-LLMA-SCD truncates long-length input by preserving texts near an examined boundary while removing those at the other sides. BF-LLMA-SCD was trained on three PAN datasets. Comparison results with the top-performing SOTA solutions show that BF-LLMA-SCD achieved the best performance results in terms of F1 on PAN 2021 and PAN 2022/D1, while obtaining competitive results on PAN 2022/D3. BF-LLMA-SCD was also trained on an Arabic SCD dataset comprising three difficulty levels. It achieved an F1 score above 0.99 on easy instances.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Authorship Attribution and Profiling · Topic Modeling · Spam and Phishing Detection
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
Department of Computer Science, King Saud University, Riyadh 11451, Saudi Arabia
Computer Science Department, Imam Mohammad Ibn Saud Islamic University, Riyadh 11564, Saudi Arabia
References (86)
Show source-provided references
1. Akiva, N., and Koppel, M. (2012, January 22–24). Identifying Distinct Components of a Multi-author Document. Proceedings of the 2012 European Intelligence and Security Informatics Conference, Odense, Denmark.10.1109/EISIC.2012.16 ↗
2. Authorship Identification of Documents with High Content Similarity10.1007/s11192-018-2661-6 ↗
3. Ensemble-Based Clustering for Writing Style Change Detection in Multi-Authored Textual Documents
4. Harnessing the Power of LLMs in Practice: A Survey on ChatGPT and Beyond
5. Dettmers, T., Pagnoni, A., Holtzman, A., and Zettlemoyer, L. (2023, January 10–16). QLoRA: Efficient Finetuning of Quantized LLMs. Proceedings of the 37th International Conference on Neural Information Processing Systems (NIPS ’23), New Orleans, LA, USA.
6. Wang, H., Riddell, A., and Juola, P. (2021, January 19–23). Mode Effects’ Challenge to Authorship Attribution. Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, Online.10.18653/v1/2021.eacl-main.97 ↗
7. Patterning of writing style evolution by means of dynamic similarity10.1016/j.patcog.2017.12.011 ↗
8. Stylometry-based Approach for Detecting Writing Style Changes in Literary Texts
9. Overview of the Author Identification Task at PAN-2017: Style Breach Detection and Author Clustering
10. Arabic Natural Language Processing: Challenges and Solutions10.1145/1644879.1644881 ↗
11. Habash, N., Soudi, A., and Buckwalter, T. (2007). On Arabic transliteration. Arabic Computational Morphology: Knowledge-Based and Empirical Methods, Springer.
12. Zitouni, I. (2014). Morphological processing of semitic languages. Natural Language Processing of Semitic Languages, Springer.10.1007/978-3-642-45358-8 ↗
13. Ahmed, M.A., and Trausan-Matu, S. (2017, January 21–23). Using natural language processing for analyzing Arabic poetry rhythm. Proceedings of the 2017 16th RoEduNet Conference: Networking in Education and Research (RoEduNet), Targu Mures, Romania.10.1109/ROEDUNET.2017.8123759 ↗
14. Writing Style Change Detection: State of the Art, Challenges, and Research Opportunities10.1007/s10462-025-11402-8 ↗
15. Style Breach Detection: An Unsupervised Detection Model
16. A Model for Style Change Detection at a Glance
17. OPI-JSA at CLEF 2017: Author Clustering and Style Breach Detection
18. Ramachandran, K.M., and Tsokos, C.P. (2020). Mathematical Statistics with Applications in R, Elsevier. [3rd ed.].
19. Writing Style Change Detection on Multi-Author Documents
20. Style Change Detection using Discourse Markers
21. Detecting a Change of Style Using Text Statistics
22. Style Change Detection using Siamese Neural Networks
23. Elamine, M., Mechti, S., and Belguith, L.H. (2019, January 3–7). An Unsupervised Method for Detecting Style Breaches in a Document. Proceedings of the 2019 IEEE/ACS 16th International Conference on Computer Systems and Applications (AICCSA), Abu Dhabi, United Arab Emirates.10.1109/AICCSA47632.2019.9035264 ↗
24. Mandic, L., Milkovic, F., and Doria, S. (2019). Combining the Powers of Clustering Affinities in Style Change Detection, University of Zagreb. Course Project Reports.
25. Style Change Detection with Feed-forward Neural Networks
26. Devlin, J., Chang, M.W., Lee, K., and Toutanova, K. (2019, January 2–7). BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. Proceedings of the the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Minneapolis, MN, USA.
27. Liu, Y., Ott, M., Goyal, N., Du, J., Joshi, M., Chen, D., Levy, O., Lewis, M., Zettlemoyer, L., and Stoyanov, V. (2019). RoBERTa: A Robustly Optimized BERT Pretraining Approach. arXiv.
28. Lan, Z., Chen, M., Goodman, S., Gimpel, K., Sharma, P., and Soricut, R. (2020, January 26–30). ALBERT: A Lite BERT for Self-supervised Learning of Language Representations. Proceedings of the International Conference on Learning Representations, Addis Ababa, Ethiopia.
29. Clark, K., Luong, M.T., Le, Q.V., and Manning, C.D. (May, January 26). ELECTRA: Pre-training Text Encoders as Discriminators Rather Than Generators. Proceedings of the 8th International Conference on Learning Representations, Addis Ababa, Ethiopia.
30. He, P., Liu, X., Gao, J., and Chen, W. (2021, January 3–7). DeBERTa: Decoding-enhanced BERT with Disentangled Attention. Proceedings of the International Conference on Learning Representations ICLR 2021, Vienna, Austria.
31. Muennighoff, N., Wang, T., Sutawika, L., Roberts, A., Biderman, S., Le Scao, T., Bari, M.S., Shen, S., Yong, Z.X., and Schoelkopf, H. (2023, January 9–14). Crosslingual Generalization through Multitask Finetuning. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics, Toronto, ON, Canada.10.18653/v1/2023.acl-long.891 ↗
32. Zhang, Z., Han, X., Liu, Z., Jiang, X., Sun, M., and Liu, Q. (August, January 28). ERNIE: Enhanced Language Representation with Informative Entities. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, Florence, Italy.
33. Understanding writing style in social media with a supervised contrastively pre-trained transformer
34. Reimers, N., and Gurevych, I. (2019, January 3–7). Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), Hong Kong, China.10.18653/v1/D19-1410 ↗
35. Style Change Detection Based On Bert And Conv1d
36. Style Change Detection based on Prompt
37. Style Change Detection Based on Bi-LSTM And Bert
38. Style Change Detection Based On Writing Style Similarity
39. Ensemble Pre-trained Transformer Models for Writing Style Change Detection
40. Team foshan-university-of-guangdong at PAN: Adaptive Entropy-Based Stability-Plasticity for Multi-Author Writing Style Analysis
41. BERT-Based Similarity Measures Oriented Approach for Style Change Detection
42. Style Change Detection: Method Based On Pre-trained Model And Similarity Recognition
43. Chen, H., Han, Z., Li, Z., and Han, Y. (2023). A Writing Style Embedding Based on Contrastive Learning for Multi-Author Writing Style Analysis. Working Notes of CLEF 2023—Conference and Labs of the Evaluation Forum, CEUR-WS.org. CEUR Workshop Proceedings.
44. Ye, Z., Zhong, C., Qi, H., and Han, Y. (2023). Supervised Contrastive Learning for Multi-Author Writing Style Analysis. Working Notes of CLEF 2023—Conference and Labs of the Evaluation Forum, CEUR-WS.org. CEUR Workshop Proceedings.
45. Kucukkaya, I.E., Sahin, U., and Toraman, C. (2023). ARC-NLP at PAN 2023: Transition-Focused Natural Language Inference for Writing Style Detection. Proceedings of the Working Notes of CLEF 2023—Conference and Labs of the Evaluation Forum, CEUR-WS.org.
46. Hashemi, A., and Shi, W. (2023). EnhancingWriting Style Change Detection using Transformer-based Models and Data Augmentation. Working Notes of CLEF 2023—Conference and Labs of the Evaluation Forum, CEUR-WS.org. CEUR Workshop Proceedings.
47. Huang, M., Huang, Z., and Kong, L. (2023). Encoded Classifier Using Knowledge Distillation for Multi-Author Writing Style Analysis. Working Notes of CLEF 2023—Conference and Labs of the Evaluation Forum, CEUR-WS.org. CEUR Workshop Proceedings.
48. Team NYCU-NLP at PAN 2024: Integrating Transformers with Similarity Adjustments for Multi-Author Writing Style Analysis
49. Team Text Understanding and Analysis at PAN: Utilizing BERT Series Pre-training Model for Multi-Author Writing Style Analysis
50. Team bingezzzleep at PAN: A Writing Style Change Analysis Model Based on RoBERTa Encoding and Contrastive Learning for Multi-Author Writing Style Analysis
51. Team Chen at PAN: Integrating R-Drop and Pre-trained Language Model for Multi-author Writing Style Analysis
52. Team baker at PAN: Enhancing Writing Style Change Detection with Virtual Softmax
53. Team karami-sh at PAN: Transformer-based Ensemble Learning for Multi-Author Writing Style Analysis
54. Team Liuc0757 at PAN: A Writing Style Embedding Method Based on Contrastive Learning for Multi-Author Writing Style Analysis
55. Zamir, M.T., Ayub, M.A., Gul, A., Ahmad, N., and Ahmad, K. (2024). Stylometry Analysis of Multi-authored Documents for Authorship and Author Style Change Detection. arXiv.
56. Continual Transfer Learning With Progress Prompt for Multi-Author Writing Style Analysis
57. DeBERTa-v3 with R-Drop regularization for Multi-Author Writing Style Analysis
58. Team fosu-stu at PAN: Supervised Fine-Tuning of Large Language Models for Multi Author Writing Style Analysis
59. Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., and Azhar, F. (2023). LLaMA: Open and Efficient Foundation Language Models. arXiv.
60. Fine-Tuned Reasoning for Writing Style Analysis
61. GPT-3: Its Nature, Scope, Limits, and Consequences10.1007/s11023-020-09548-1 ↗
62. Exploring the limits of transfer learning with a unified text-to-text transformer
63. Character-based Convolutional Neural Network for Style Change Detection
64. A Parallel Hierarchical Attention Network for Style Change Detection
65. Alsheddi, A.S., and Menai, M.E.B. (2025). Edge Convolutional Networks for Style Change Detection in Arabic Multi-Authored Text. Appl. Sci., 15.10.3390/app15126633 ↗
66. Overview of the Style Change Detection Task at PAN 2022
67. On row rank equal column rank10.1080/00207390802357273 ↗
68. Hu, E.J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., and Chen, W. (2022, January 25–29). LoRA: Low-Rank Adaptation of Large Language Models. Proceedings of the International Conference on Learning Representations (ICLR 2022), Online.
69. Dettmers, T., and Zettlemoyer, L. (2023, January 23–29). The case for 4-bit precision: K-bit Inference Scaling Laws. Proceedings of the 40th International Conference on Machine Learning (ICML’23), Honolulu, HI, USA.
70. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A.N., Kaiser, T., and Polosukhin, I. (2017, January 4–9). Attention is All you Need. Proceedings of the 31st International Conference on Neural Information Processing Systems (NIPS’17), Long Beach, CA, USA.
71. Connor, R., Dearle, A., Claydon, B., and Vadicamo, L. (2024). Correlations of Cross-Entropy Loss in Machine Learning. Entropy, 26.10.3390/e26060491 ↗
72. Long Short-Term Memory Based Recurrent Neural Network Architectures for Large Vocabulary Speech Recognition
73. Overview of the Style Change Detection Task at PAN 2021
74. Overview of the Style Change Detection Task at PAN 2020
75. Zangerle, E., Mayerl, M., Potthast, M., and Stein, B. (2023). Overview of the Multi-Author Writing Style Analysis Task at PAN 2023. Working Notes of CLEF 2023—Conference and Labs of the Evaluation Forum, CEUR-WS.org. CEUR Workshop Proceedings.
76. Overview of the Multi-Author Writing Style Analysis Task at PAN 2024
77. Team riyahsanjesh at PAN: Multi-feature with CNN and Bi-LSTM Neural Network Approach to Style Change Detection
78. Kingma, D.P., and Ba, J. (2015, January 7–9). Adam: A Method for Stochastic Optimization. Proceedings of the International Conference on Learning Representations (ICLR 2015), San Diego, CA, USA.
79. Style Change Detection on Real-World Data using an LSTM-powered Attribution Algorithm
80. Wazrah, A.A.A., Altamimi, A., Aljasim, H., Alshammari, W., Al-Matham, R., Elnashar, O., Amin, M., and AlOsaimy, A. (2025, January 19). Evaluation of Large Language Models on Arabic Punctuation Prediction. Proceedings of the 1st Workshop on NLP for Languages Using Arabic Script, Abu Dhabi, United Arab Emirates.
81. Multi-label Style Change Detection by Solving a Binary Classification Problem
82. Three Style Similarity: Sentence-embedding, Auxiliary Words, Punctuation
83. Grattafiori, A., Dubey, A., Jauhri, A., Pandey, A., Kadian, A., Al-Dahle, A., Letman, A., Mathur, A., Schelten, A., and Vaughan, A. (2024). The Llama 3 Herd of Models. arXiv.
84. Unlocking the black box: An in-depth review on interpretability, explainability, and reliability in deep learning10.1007/s00521-024-10437-2 ↗
85. Hung, C.Y., Hu, Z., Hu, Y., and Lee, R. (2023, January 6–10). Who Wrote it and Why? Prompting Large-Language Models for Authorship Verification. Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2023, Singapore.10.18653/v1/2023.findings-emnlp.937 ↗
86. Huang, B., Chen, C., and Shu, K. (2024, January 12–16). Can Large Language Models Identify Authorship?. Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2024, Miami, FL, USA.10.18653/v1/2024.findings-emnlp.26 ↗