Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data

@article{jiapeng2026,
  title = {Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data},
  author = {Wang, Jiapeng and Hu, Yiwen and Gao, Yanzipeng and Wang, Haoyu and Wang, Shuo and Lu, Hongyu and Mao, Jiaxin and Zhao, Xin and Li, Junyi and Zhang, Xiao},
  journal = {Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing},
  year = {2026},
  month = {Jan},
}

Wang, Jiapeng et al. (2026). Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data. Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing.

Wang, Jiapeng, et al.. "Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data." Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, 2026.

Wang, Jiapeng, et al.. "Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data." Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing (2026).

Wang, Jiapeng et al., 2026. Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data. Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing.

TY  - JOUR
TI  - Entropy-Guided Token Dropout: Training Autoregressive Language Models with Limited Domain Data
AU  - Wang, Jiapeng
AU  - Hu, Yiwen
AU  - Gao, Yanzipeng
AU  - Wang, Haoyu
AU  - Wang, Shuo
AU  - Lu, Hongyu
AU  - Mao, Jiaxin
AU  - Zhao, Xin
AU  - Li, Junyi
AU  - Zhang, Xiao
PY  - 2026
JO  - Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
ER  -