---
title: "Decoding Text Spans for Efficient and Accurate Named-Entity"
description: "The authors, from Samsung Research UK, propose SpanDec. 1.8 points. 26 references. Among them: 0."
authors: ["Andrea Maracani", "Savas Ozkan", "Junyi Zhu", "Sinan Mutlu", "Mete Ozay"]
year: 2026
venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track)"
source: "https://aclanthology.org/2026.acl-industry.71/"
doi: "10.18653/v1/2026.acl-industry.71"
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/maracani-2026-decoding-text-spans-efficient-accurate/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/maracani-2026-decoding-text-spans-efficient-accurate/)

[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/27/) / Decoding Text Spans for Efficient and Accurate Named-Entity Recognition

# Decoding Text Spans for Efficient and Accurate Named-Entity Recognition

**Authors:** Andrea Maracani, Savas Ozkan, Junyi Zhu, Sinan Mutlu and [Mete Ozay](https://md.research.elamur.ai/free/en/authors/mete-ozay/) · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/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.acl-industry.71/) · [PDF](https://aclanthology.org/2026.acl-industry.71.pdf) · [DOI 10.18653/v1/2026.acl-industry.71](https://doi.org/10.18653/v1/2026.acl-industry.71) · [OpenAlex](https://openalex.org/W7166890525)

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

## Summary
Span-based named-entity recognition, especially the PL-Marker family, tends to beat plain token tagging on accuracy. Its cost is high, though: it enumerates many candidate spans and pushes marker tokens through every encoder layer. That makes it expensive for high-volume or on-device serving.

The authors, from Samsung Research UK, propose SpanDec. It replaces the last encoder layer with a lightweight decoder that is used only for span markers. The text is encoded once, and span–marker interaction happens only at the final stage, so the total parameter count stays roughly the same. A second variant, SF-SpanDec, adds a small binary classifier that prunes tokens unlikely to belong to any entity before spans are decoded.

They tested the approach on CoNLL++, CrossNER, OntoNotes5 and BC5CDR, using MiniLM, BERT-Base and RoBERTa-Large encoders. SpanDec matches or slightly exceeds PL-Marker's F1 while running much faster and using far fewer GFLOPs. SF-SpanDec comes close to token-classification speed while keeping most of the accuracy gain. This makes span-level NER practical for latency-bound production pipelines.

## Key points
- Against PL-Marker, the method raises throughput by up to 2.7× and cuts GFLOPs by up to 8.2×. Its average F1 is 1.8 points higher than the token-classification baseline.
- With MiniLM, average F1 and throughput (samples/s) are:
  - token classification: 84.7 at 330.8
  - PL-Marker: 85.4 at 140.3
  - SpanDec: 86.2 at 241.3
  - SF-SpanDec: 85.8 at 310.8
- Span filtering typically keeps only about 15% of candidate spans. As a result, SF-SpanDec needs roughly the same GFLOPs as token classification and reaches 92–93% of its throughput.
- With RoBERTa-L, SpanDec scores 89.4 average F1 against 89.5 for PL-Marker, and its throughput is 118.7 versus 68.0 samples/s.
- The paper also compares against SOTA systems (GLiNER, UniNER, InstructUIE, GPT-NER, SplitNER and 1-shot LLMs). There, SpanDec with 0.35B parameters achieves F1 of 90.8 on BC5CDR, 95.4 on CoNLL and 91.3 on OntoNotes, with higher throughput than all the compared systems.

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

📚 Shelves (5): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 16,667 works · [🤖 Artificial Intelligence](https://md.research.elamur.ai/free/en/shelves/computer-science/artificial-intelligence/) · 16,194 works · [🎭 Arts and Humanities](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/) · 16,039 works · [🗣️ Language and Linguistics](https://md.research.elamur.ai/free/en/shelves/arts-and-humanities/language-and-linguistics/) · 15,891 works · [👁️ Computer Vision and Pattern Recognition](https://md.research.elamur.ai/free/en/shelves/computer-science/computer-vision-and-pattern-recognition/) · 2,869 works

🧭 Topics (1): [🏷️ Named entity recognition](https://md.research.elamur.ai/free/en/topics/named-entity-recognition/) · 200 works

## Cite

Cite as: Andrea Maracani, Savas Ozkan, Junyi Zhu, Sinan Mutlu and Mete Ozay. 2026. Decoding Text Spans for Efficient and Accurate Named-Entity Recognition. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track).

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

```bibtex
@inproceedings{maracani2026decoding,
  title = {{Decoding Text Spans for Efficient and Accurate Named-Entity Recognition}},
  author = {Andrea Maracani, Savas Ozkan, Junyi Zhu, Sinan Mutlu and Mete Ozay},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-industry.71},
  url = {https://doi.org/10.18653/v1/2026.acl-industry.71}
}
```

- [← previous Debiasing Reward Models via Causally Motivated Inference-Time Intervention](https://md.research.elamur.ai/free/en/papers/shinoda-2026-debiasing-reward-models-causally-motivated/)
- [next → Decoding-Unlearning: Fact Forgetting via Entropy-Guided Inference](https://md.research.elamur.ai/free/en/papers/pu-2026-decoding-unlearning-fact-forgetting-entropy/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/27/)

## References

26 references, 10 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Tom Aarsen. 2023. Spanmarker for named entity recognition. [https://github.com/tomaarsen/](https://github.com/tomaarsen/) SpanMarkerNER. [↗](https://github.com/tomaarsen/)
2. Arjun Ramesh Binu Rajan M. R. Anupama M. Nair, Anusha Aji Justus. 2020. Event extraction from emails. International Journal of Computer Applications, 176(41):1–8. [find](https://scholar.google.com/scholar?q=Arjun+Ramesh+Binu+Rajan+M%2E+R%2E+Anupama+M%2E+Nair%2C+Anusha+Aji+Justus%2E+2020%2E+Event+extraction+from+emails%2E+International+Journal+of+Computer+Applications%2C+176%2841%29%3A1%E2%80%938%2E)
3. Jatin Arora and Youngja Park. 2023. Split-NER: Named entity recognition via two question-answering-based classifications. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). [find](https://scholar.google.com/scholar?q=Jatin+Arora+and+Youngja+Park%2E+2023%2E+Split%2DNER%3A+Named+entity+recognition+via+two+question%2Danswering%2Dbased+classifications%2E+In+Proceedings+of+the+61st+Annual+Meeting+of+the+Association+for+Computational)
4. Sergei Bogdanov, Alexandre Constantin, Timothée Bernard, Benoit Crabbé, and Etienne P Bernard. 2024. NuNER: Entity recognition encoder pre-training via LLM-annotated data. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 11829–11841, Miami, Florida, US… [find](https://scholar.google.com/scholar?q=Sergei+Bogdanov%2C+Alexandre+Constantin%2C+Timoth%C3%A9e+Bernard%2C+Benoit+Crabb%C3%A9%2C+and+Etienne+P+Bernard%2E+2024%2E+NuNER%3A+Entity+recognition+encoder+pre%2Dtraining+via+LLM%2Dannotated+data%2E+In+Proceedings+of+the+2024+C)
5. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. [BERT: Pre-training of deep bidirectional transformers for language understanding](https://md.research.elamur.ai/free/en/papers/devlin-2019-bert-pre-training-deep-bidirectional/). In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,…
6. Christopher Dozier, Ravikumar Kondadadi, Marc Light, Arun Vachher, Sriharsha Veeramachaneni, and Ramdev Wudali. 2010. Named Entity Recognition and Resolution in Legal Text, pages 27–43. Springer Berlin Heidelberg, Berlin, Heidelberg. [find](https://scholar.google.com/scholar?q=Christopher+Dozier%2C+Ravikumar+Kondadadi%2C+Marc+Light%2C+Arun+Vachher%2C+Sriharsha+Veeramachaneni%2C+and+Ramdev+Wudali%2E+2010%2E+Named+Entity+Recognition+and+Resolution+in+Legal+Text%2C+pages+27%E2%80%9343%2E+Springer+Berli)
7. Sumam Francis, Jordy Van Landeghem, and Marie-Francine Moens. 2019. Transfer learning for named entity recognition in financial and biomedical documents. Information, 10(8). [find](https://scholar.google.com/scholar?q=Sumam+Francis%2C+Jordy+Van+Landeghem%2C+and+Marie%2DFrancine+Moens%2E+2019%2E+Transfer+learning+for+named+entity+recognition+in+financial+and+biomedical+documents%2E+Information%2C+10%288%29%2E)
8. Eduard Hovy, Mitchell Marcus, Martha Palmer, Lance Ramshaw, and Ralph Weischedel. 2006. OntoNotes: The 90% solution. In Proceedings of the Human Language Technology Conference of the NAACL, Companion Volume: Short Papers, pages 57–60, New York City, USA. Association for Computational Linguistics. [find](https://scholar.google.com/scholar?q=Eduard+Hovy%2C+Mitchell+Marcus%2C+Martha+Palmer%2C+Lance+Ramshaw%2C+and+Ralph+Weischedel%2E+2006%2E+OntoNotes%3A+The+90%25+solution%2E+In+Proceedings+of+the+Human+Language+Technology+Conference+of+the+NAACL%2C+Companion+)
9. Seoyeon Kim, Kwangwook Seo, Hyungjoo Chae, Jinyoung Yeo, and Dongha Lee. 2024. VerifiNER: Verification-augmented NER via knowledge-grounded reasoning with large language models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). [find](https://scholar.google.com/scholar?q=Seoyeon+Kim%2C+Kwangwook+Seo%2C+Hyungjoo+Chae%2C+Jinyoung+Yeo%2C+and+Dongha+Lee%2E+2024%2E+VerifiNER%3A+Verification%2Daugmented+NER+via+knowledge%2Dgrounded+reasoning+with+large+language+models%2E+In+Proceedings+of+the+)
10. Jiao Li, Yueping Sun, Robin J. Johnson, Daniela Sci-aky, Chih-Hsuan Wei, Robert Leaman, Allan Peter Davis, Carolyn J. Mattingly, Thomas C. Wiegers, and Zhiyong Lu. 2016. Biocreative V CDR task corpus: a resource for chemical disease relation extraction. [find](https://scholar.google.com/scholar?q=Jiao+Li%2C+Yueping+Sun%2C+Robin+J%2E+Johnson%2C+Daniela+Sci%2Daky%2C+Chih%2DHsuan+Wei%2C+Robert+Leaman%2C+Allan+Peter+Davis%2C+Carolyn+J%2E+Mattingly%2C+Thomas+C%2E+Wiegers%2C+and+Zhiyong+Lu%2E+2016%2E+Biocreative+V+CDR+task+corpus%3A)
11. Database J. Biol. Databases Curation, 2016. [find](https://scholar.google.com/scholar?q=Database+J%2E+Biol%2E+Databases+Curation%2C+2016%2E)
12. Zihan Liu, Yan Xu, Tiezheng Yu, Wenliang Dai, Ziwei Ji, Samuel Cahyawijaya, Andrea Madotto, and Pascale Fung. 2020. Crossner: Evaluating cross-domain named entity recognition. [find](https://scholar.google.com/scholar?q=Zihan+Liu%2C+Yan+Xu%2C+Tiezheng+Yu%2C+Wenliang+Dai%2C+Ziwei+Ji%2C+Samuel+Cahyawijaya%2C+Andrea+Madotto%2C+and+Pascale+Fung%2E+2020%2E+Crossner%3A+Evaluating+cross%2Ddomain+named+entity+recognition%2E)
13. Hiroki Nakayama. 2018. seqeval: A python framework for sequence labeling evaluation. Software available from [https://github.com/chakki-works/seqeval](https://github.com/chakki-works/seqeval). [↗](https://github.com/chakki-works/seqeval)
14. Kim Sang Tjong, F. Erik, and Fien De Meulder. 2003. [Introduction to the CoNLL-2003 shared task: Language-independent named entity recognition](https://md.research.elamur.ai/free/en/papers/sang-2003-introduction-conll-2003-shared-task/). In Proceedings of the Seventh Conference on Natural Language Learning at HLT-NAACL 2003, pages 142– 147.
15. David Wadden, Ulme Wennberg, Yi Luan, and Hannaneh Hajishirzi. 2019. [Entity, relation, and event extraction with contextualized span representations](https://md.research.elamur.ai/free/en/papers/wadden-2019-entity-relation-event-extraction-contextualized/). In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Langua…
16. Shuhe Wang, Xiaofei Sun, Xiaoya Li, Rongbin Ouyang, Fei Wu, Tianwei Zhang, Jiwei Li, Guoyin Wang, and Chen Guo. 2025. [GPT-NER: Named entity recognition via large language models](https://md.research.elamur.ai/free/en/papers/wang-2025-gpt-ner-named-entity-recognition/). In Findings of the Association for Computational Linguistics: NAACL 2025, pages 4257–4275, Albuquerque, New Mexico. Assoc…
17. Wenhui Wang, Furu Wei, Li Dong, Hangbo Bao, Nan Yang, and Ming Zhou. 2020. Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers. In Advances in Neural Information Processing Systems (NeurIPS). [find](https://scholar.google.com/scholar?q=Wenhui+Wang%2C+Furu+Wei%2C+Li+Dong%2C+Hangbo+Bao%2C+Nan+Yang%2C+and+Ming+Zhou%2E+2020%2E+Minilm%3A+Deep+self%2Dattention+distillation+for+task%2Dagnostic+compression+of+pre%2Dtrained+transformers%2E+In+Advances+in+Neural+Inf)
18. Xingyao Wang, Wenxuan Zhou, Chao Zu, Haoran Xia, Tian Chen, Yong Zhang, Rui Zheng, Jing Ye, Qianying Zhang, Tao Gui, Jian Kang, Junfeng Yang, Sha Li, and Chao Du. 2023. [Multi-task instruction tuning for unified information extraction](https://md.research.elamur.ai/free/en/papers/wang-2023-instructuie-multi-task-instruction-tuning/). Preprint, arXiv:2304.08085.
19. Xinyu Wang, Yong Jiang, Nguyen Bach, Tao Wang, Zhongqiang Huang, Fei Huang, and Kewei Tu. 2021. [Automated concatenation of embeddings for structured prediction](https://md.research.elamur.ai/free/en/papers/wang-2021-automated-concatenation-embeddings-structured-prediction/). In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on N…
20. Zihan Wang, Jingbo Shang, Liyuan Liu, Lihao Lu, Jiacheng Liu, and Jiawei Han. 2019. Crossweigh: Training named entity tagger from imperfect annotations. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Lan… [find](https://scholar.google.com/scholar?q=Zihan+Wang%2C+Jingbo+Shang%2C+Liyuan+Liu%2C+Lihao+Lu%2C+Jiacheng+Liu%2C+and+Jiawei+Han%2E+2019%2E+Crossweigh%3A+Training+named+entity+tagger+from+imperfect+annotations%2E+In+Proceedings+of+the+2019+Conference+on+Empiri)
21. Thomas Wolf, Lysandre Debut, Victor Sanh, and 1 others. 2020. [Transformers: State-of-the-art natural language processing](https://md.research.elamur.ai/free/en/papers/wolf-2020-transformers-state-art-natural-language/). In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, pages 38–45. Association for Computational Linguistics.
22. Deming Ye, Yankai Lin, Peng Li, and Maosong Sun. 2022. [Packed levitated marker for entity and relation extraction](https://md.research.elamur.ai/free/en/papers/ye-2022-packed-levitated-marker-entity-relation/). In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4904–4917, Dublin, Ireland. Association for Computational Lingu…
23. Urchade Zaratiana and 1 others. 2024. [GLiNER: Generalist model for named entity recognition using bidirectional transformer](https://md.research.elamur.ai/free/en/papers/zaratiana-2024-gliner-generalist-model-named-entity/). In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics (NAACL).
24. Zexuan Zhong and Danqi Chen. 2021. [A frustratingly easy approach for entity and relation extraction](https://md.research.elamur.ai/free/en/papers/zhong-2020-frustratingly-easy-approach-entity-relation/). In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 50–61, Online. Association for Computational L…
25. Wenxuan Zhou, Sheng Zhang, Yu Gu, Muhao Chen, and Hoifung Poon. 2023. Universalner: Targeted distillation from large language models for open named entity recognition. arXiv preprint arXiv:2308.03279. [find](https://scholar.google.com/scholar?q=Wenxuan+Zhou%2C+Sheng+Zhang%2C+Yu+Gu%2C+Muhao+Chen%2C+and+Hoifung+Poon%2E+2023%2E+Universalner%3A+Targeted+distillation+from+large+language+models+for+open+named+entity+recognition%2E+arXiv+preprint+arXiv%3A2308%2E03279%2E)
26. Dengya Zhu, Sirui Li, Nik Thompson, and Kok Wai Wong. 2024. Open-source large language models excel in named entity recognition. In International Conference on Neural Information Processing, pages 313–326. Springer. [find](https://scholar.google.com/scholar?q=Dengya+Zhu%2C+Sirui+Li%2C+Nik+Thompson%2C+and+Kok+Wai+Wong%2E+2024%2E+Open%2Dsource+large+language+models+excel+in+named+entity+recognition%2E+In+International+Conference+on+Neural+Information+Processing%2C+pages+313)

*Updated 1 November 2026*