Proceedings

TS-PEFT: Unveiling Token-Level Redundancy in Parameter-Efficient Fine-Tuning

Dabiao Ma, Ziming Dai, Zhimin Xin, Shu Wang, Jian Yang, Haojun Fei

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 4616-4624.

https://doi.org/10.24963/ijcai.2026/514

Keywords

  • Natural Language Processing: Interpretability and analysis of models for NLP
  • Machine Learning: Learning sparse models
  • Machine Learning: Multi-task and transfer learning

Abstract

Current Parameter-Efficient Fine-Tuning (PEFT) methods typically operate under an implicit assumption: once a target module is selected, every token passing through it contributes equally to the downstream task and requires a parameter update. In this paper, we challenge this convention by revealing a pervasive token-level redundancy in the fine-tuning of large models (LMs). We propose TS-PEFT, a theoretical framework utilizing proximal optimization that acts as a dynamic probe to identify token-level redundancy during the fine-tuning process. Extensive experiments demonstrate that indiscriminately updating all tokens is not only computationally superfluous but often introduces optimization noise. Surprisingly, by discarding 30%-70% of token updates, TS-PEFT consistently matches or exceeds the performance of dense baselines such as LoRA, DoRA. Our in-depth analysis shows that the learned token-level sparsity is a superior indicator of module importance compared to traditional weight criteria, providing a novel data-driven perspective on the intrinsic adaptation mechanism of LMs.

Dabiao Ma, Ziming Dai, Zhimin Xin, Shu Wang, Jian Yang, and Haojun Fei. 2026. TS-PEFT: Unveiling Token-Level Redundancy in Parameter-Efficient Fine-Tuning. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 4616–4624. https://doi.org/10.24963/ijcai.2026/514
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards