---
title: "From “Thinking” to “Justifying” — paper by Qian et al. (2026)"
description: "They propose a \"Result → Justify\" paradigm in which the model first states its answer and then defends… 27 references, cited by 1 paper here."
authors: ["Chen Qian", "Yimeng Wang", "Yu Chen", "Lingfei Wu", "Andreas Stathopoulos"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.1232/"
doi: "10.18653/v1/2026.findings-acl.1232"
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/qian-2026-thinking-justifying-aligning-high-stakes/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/qian-2026-thinking-justifying-aligning-high-stakes/)

[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/13/) / From “Thinking” to “Justifying”: Aligning High-Stakes Explainability with Professional Communication Standards

# From “Thinking” to “Justifying”: Aligning High-Stakes Explainability with Professional Communication Standards

**Authors:** [Chen Qian](https://md.research.elamur.ai/free/en/authors/chen-qian/), Yimeng Wang, [Yu Chen](https://md.research.elamur.ai/free/en/authors/yu-chen/), [Lingfei Wu](https://md.research.elamur.ai/free/en/authors/lingfei-wu/) and Andreas Stathopoulos · **Year:** 2026 · **Venue:** [Findings of the Association for Computational Linguistics: ACL 2026](https://md.research.elamur.ai/free/en/venues/findings/) ([Findings of the Association for Computational Linguistics 2026](https://md.research.elamur.ai/free/en/venues/findings/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.findings-acl.1232/) · [PDF](https://aclanthology.org/2026.findings-acl.1232.pdf) · [DOI 10.18653/v1/2026.findings-acl.1232](https://doi.org/10.18653/v1/2026.findings-acl.1232) · [OpenAlex](https://openalex.org/W7166835541)

**Citations:** 0 citations (OpenAlex, 2026-10-14) · 1 in our library, of the 149,439 papers whose bibliographies we read

## Summary
The paper argues that Chain-of-Thought explanations in high-stakes settings are hard to verify. Because the model reasons before it answers, the reasoning can drift or contain gaps, and the final answer may not match the rationale. The authors borrow from professional writing conventions, namely CREAC from legal writing and BLUF from military and business communication. They propose a "Result → Justify" paradigm in which the model first states its answer and then defends it in a fixed structure.

Their SEF (Structured Explainability Framework) has two uses:
- **Evaluation:** a set of six heuristic metrics, split into plausibility (structure) and faithfulness (evidence grounding).
- **Generation:** a four-step prompt (Commit, Ground, Defend, Conclude).

The authors tested SEF on four open-weight 12–14B models across four binary legal, medical and financial tasks. Every metric correlated with answer correctness, and SEF outperformed all six prompting baselines on average accuracy.

This matters because it offers a cheap, scalable way to produce explanations that reviewers can check section by section. Human expert evaluation is still left for future work.

## Key points
- **Metrics:** AFL, AC and CI cover plausibility, while DTC, CEA and FS cover faithfulness. Computed on 90,608 outputs from all methods, their Pearson r with correctness ranged from 0.20 to 0.42 (p \< 0.001). Plausibility metrics correlated more strongly than faithfulness metrics.
- **Accuracy:** averaged over models, SEF reached 83.9%, compared with 78.6% for CoT (+5.3 points). Direct answering (80.9%) beat CoT, and ToT, CoVe, Vanilla RAG and Self-RAG all scored lower than SEF.
- **Data:** the tasks were Financial PhraseBank, ConsumerQA, Hearsay (LegalBench) and PubMedQA, totalling 1,618 test samples.
- **Models:** DeepSeek-R1-Distill-Qwen-14B, Gemma 3 12B, Ministral-3-14B and Qwen 2.5 14B, all run with greedy decoding via vLLM.
- **Ablations:** removing the plausibility instructions cut accuracy by 32.9 points, while removing the faithfulness instructions cut it by 4.3. DTC mattered most for Hearsay (−6.2). Code is available at github.com/cqian03/SEF.

*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 (6): [🖥️ 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 · [🤝 Social Sciences](https://md.research.elamur.ai/free/en/shelves/social-sciences/) · 1,772 works · [🤝 Safety Research](https://md.research.elamur.ai/free/en/shelves/social-sciences/safety-research/) · 543 works

🧭 Topics (1): [🔬 Interpretability](https://md.research.elamur.ai/free/en/topics/interpretability/) · 130 works

## Cite

Cite as: Chen Qian, Yimeng Wang, Yu Chen, Lingfei Wu and Andreas Stathopoulos. 2026. From “Thinking” to “Justifying”: Aligning High-Stakes Explainability with Professional Communication Standards. Findings of the Association for Computational Linguistics: ACL 2026.

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

```bibtex
@inproceedings{qian2026thinking,
  title = {{From “Thinking” to “Justifying”: Aligning High-Stakes Explainability with Professional Communication Standards}},
  author = {Chen Qian, Yimeng Wang, Yu Chen, Lingfei Wu and Andreas Stathopoulos},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1232},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1232}
}
```

- [← previous From Weights to Activations: Is Steering the Next Frontier of Adaptation?](https://md.research.elamur.ai/free/en/papers/ostermann-2026-weights-activations-is-steering-next/)
- [next → FS-Researcher: Test-Time Scaling for Long-Horizon Research Tasks with File-System-Based Agents](https://md.research.elamur.ai/free/en/papers/zhu-2026-fs-researcher-test-time-scaling/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/13/)

## References

27 references, 17 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Martin Aigner and Günter M. Ziegler. 2010. Proofs from THE BOOK. Springer Berlin Heidelberg. [find](https://scholar.google.com/scholar?q=Martin+Aigner+and+G%C3%BCnter+M%2E+Ziegler%2E+2010%2E+Proofs+from+THE+BOOK%2E+Springer+Berlin+Heidelberg%2E)
2. Akari Asai, Zeqiu Wu, Yizhong Wang, Avirup Sil, and Hannaneh Hajishirzi. 2023. [Self-rag: Learning to retrieve, generate, and critique through self-reflection](https://md.research.elamur.ai/free/en/papers/asai-2023-self-rag-learning-retrieve-generate/). arXiv preprint.
3. Jinze Bai, Shuai Bai, Yunfei Chu, Zeyu Cui, Kai Dang, Xiaodong Deng, Yang Fan, Wenbin Ge, Yu Han, Fei Huang, Binyuan Hui, Luo Ji, Mei Li, Junyang Lin, Runji Lin, Dayiheng Liu, Gao Liu, Chengqiang Lu, Keming Lu, and 29 others. 2023. [Qwen technical report](https://md.research.elamur.ai/free/en/papers/bai-2023-qwen-technical-report/). arXiv preprint.
4. Rishi Bommasani, Drew A. Hudson, Ehsan Adeli, Russ Altman, Simran Arora, Sydney von Arx, Michael S. Bernstein, Jeannette Bohg, Antoine Bosselut, Emma Brunskill, Erik Brynjolfsson, Shyamal Buch, Dallas Card, Rodrigo Castellon, Niladri Chatterji, Annie Chen, Kathleen Creel, Jared Quincy Davis, Dora De… — [On the Opportunities and Risks of Foundation Models](https://md.research.elamur.ai/free/en/papers/bommasani-2021-opportunities-risks-foundation-models/)
5. Steven J Burton. 2007. An introduction to law and legal reasoning. Wolters Kluwer Law & Business. [find](https://scholar.google.com/scholar?q=Steven+J+Burton%2E+2007%2E+An+introduction+to+law+and+legal+reasoning%2E+Wolters+Kluwer+Law+%26+Business%2E)
6. DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, and 181 others. 2025. [Deepseek-r1: Incentivizing reasoning capability in ll…](https://md.research.elamur.ai/free/en/papers/deepseek-ai-2025-deepseek-r1-incentivizing-reasoning-capability/)
7. Shehzaad Dhuliawala, Mojtaba Komeili, Jing Xu, Roberta Raileanu, Xian Li, Asli Celikyilmaz, and Jason Weston. 2024. [Chain-of-verification reduces hallucination in large language models](https://md.research.elamur.ai/free/en/papers/dhuliawala-2023-chain-verification-reduces-hallucination-large/). In Findings of the Association for Computational Linguistics ACL 2024, page 3563–3578. Association for Computation…
8. Gemma Team, Thomas Mesnard, Cassidy Hardin, Robert Dadashi, Surya Bhupatiraju, Shreya Pathak, Laurent Sifre, Morgane Rivière, Mihir Sanjay Kale, Juliette Love, Pouya Tafti, Léonard Hussenot, Pier Giuseppe Sessa, Aakanksha Chowdhery, Adam Roberts, Aditya Barua, Alex Botev, Alex Castro-Ros, Ambrose Sl… [find](https://scholar.google.com/scholar?q=Gemma+Team%2C+Thomas+Mesnard%2C+Cassidy+Hardin%2C+Robert+Dadashi%2C+Surya+Bhupatiraju%2C+Shreya+Pathak%2C+Laurent+Sifre%2C+Morgane+Rivi%C3%A8re%2C+Mihir+Sanjay+Kale%2C+Juliette+Love%2C+Pouya+Tafti%2C+L%C3%A9onard+Hussenot%2C+Pier+Gius)
9. Neel Guha, Julian Nyarko, Daniel E. Ho, Christopher Ré, Adam Chilton, Aditya Narayana, Alex Chohlas-Wood, Austin Peters, Brandon Waldon, Daniel N. Rockmore, Diego Zambrano, Dmitry Talisman, Enam Hoque, Faiz Surani, Frank Fagan, Galit Sarfaty, Gregory M. Dickinson, Haggai Porat, Jason Hegland, and 21… — [LegalBench: A Collaboratively Built Benchmark for Measuring Legal Reasoning in Large Language Models](https://md.research.elamur.ai/free/en/papers/guha-2023-legalbench-collaboratively-built-benchmark-measuring/)
10. Alon Jacovi and Yoav Goldberg. 2020. [Towards faithfully interpretable nlp systems: How should we define and evaluate faithfulness?](https://md.research.elamur.ai/free/en/papers/jacovi-2020-faithfully-interpretable-nlp-systems-how/) arXiv preprint.
11. Qiao Jin, Bhuwan Dhingra, Zhengping Liu, William Cohen, and Xinghua Lu. 2019. [Pubmedqa: A dataset for biomedical research question answering](https://md.research.elamur.ai/free/en/papers/jin-2019-pubmedqa-dataset-biomedical-research-question/). In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Proce…
12. Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph Gonzalez, Hao Zhang, and Ion Stoica. 2023. [Efficient memory management for large language model serving with pagedattention](https://md.research.elamur.ai/free/en/papers/kwon-2023-efficient-memory-management-large-language/). In Proceedings of the 29th Symposium on Operating Systems Principles, SOSP ’23, page 611…
13. Leslie Lamport. 2012. How to write a 21st century proof. Journal of Fixed Point Theory and Applications, 11(1):43–63. [find](https://scholar.google.com/scholar?q=Leslie+Lamport%2E+2012%2E+How+to+write+a+21st+century+proof%2E+Journal+of+Fixed+Point+Theory+and+Applications%2C+11%281%29%3A43%E2%80%9363%2E)
14. Tamera Lanham, Anna Chen, Ansh Radhakrishnan, Benoit Steiner, Carson Denison, Danny Hernandez, Dustin Li, Esin Durmus, Evan Hubinger, Jackson Kernion, Kamilė Lukošiūtė, Karina Nguyen, Newton Cheng, Nicholas Joseph, Nicholas Schiefer, Oliver Rausch, Robin Larson, Sam McCandlish, Sandipan Kundu, an… — [Measuring Faithfulness in Chain-of-Thought Reasoning](https://md.research.elamur.ai/free/en/papers/lanham-2023-measuring-faithfulness-chain-thought-reasoning/)
15. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2020. [Retrieval-augmented generation for knowledge-intensive nlp tasks](https://md.research.elamur.ai/free/en/papers/lewis-2020-retrieval-augmented-generation-knowledge-intensive/). In Proceedings of the 34th Internation…
16. Scott M. Lundberg and Su-In Lee. 2017. [A unified approach to interpreting model predictions](https://md.research.elamur.ai/free/en/papers/lundberg-2017-unified-approach-interpreting-model-predictions/). In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, page 4768–4777, Red Hook, NY, USA. Curran Associates Inc.
17. Qing Lyu, Shreya Havaldar, Adam Stein, Li Zhang, Delip Rao, Eric Wong, Marianna Apidianaki, and Chris Callison-Burch. 2023. [Faithful chain-of-thought reasoning](https://md.research.elamur.ai/free/en/papers/lyu-2023-faithful-chain-thought-reasoning/). In Proceedings of the 13th International Joint Conference on Natural Language Processing and the 3rd Conference of the Asia-Pacific Chapter…
18. Pekka Malo, Ankur Sinha, Pekka Korhonen, Jyrki Wallenius, and Pyry Takala. 2013. Good debt or bad debt: Detecting semantic orientations in economic texts. Journal of the Association for Information Science and Technology, 65(4):782–796. [find](https://scholar.google.com/scholar?q=Pekka+Malo%2C+Ankur+Sinha%2C+Pekka+Korhonen%2C+Jyrki+Wallenius%2C+and+Pyry+Takala%2E+2013%2E+Good+debt+or+bad+debt%3A+Detecting+semantic+orientations+in+economic+texts%2E+Journal+of+the+Association+for+Information+Sc)
19. Jean Mangan, Brittany Blanchard, Gabrielle Gravel, Chase Lyndale, and Connely Doizé. 2022. 1.10: CREAC legal writing paradigm. In Legal Writing Manual. University System of Georgia via Affordable Learning Georgia. LibreTexts. [find](https://scholar.google.com/scholar?q=Jean+Mangan%2C+Brittany+Blanchard%2C+Gabrielle+Gravel%2C+Chase+Lyndale%2C+and+Connely+Doiz%C3%A9%2E+2022%2E+1%2E10%3A+CREAC+legal+writing+paradigm%2E+In+Legal+Writing+Manual%2E+University+System+of+Georgia+via+Affordable+Lear)
20. Mistral AI. 2025. Introducing mistral 3. Mistral AI Technical Blog. [find](https://scholar.google.com/scholar?q=Mistral+AI%2E+2025%2E+Introducing+mistral+3%2E+Mistral+AI+Technical+Blog%2E)
21. Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. 2016. ["why should i trust you?": Explaining the predictions of any classifier](https://md.research.elamur.ai/free/en/papers/ribeiro-2016-why-should-i-trust-explaining/). In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD ’16, page 1135–1144, New York, NY, USA. Association for Co…
22. Kabir Sehgal. 2016. How to write email with military precision. Harvard Business Review. [find](https://scholar.google.com/scholar?q=Kabir+Sehgal%2E+2016%2E+How+to+write+email+with+military+precision%2E+Harvard+Business+Review%2E)
23. Miles Turpin, Julian Michael, Ethan Perez, and Samuel R. Bowman. 2023. [Language models don’t always say what they think: unfaithful explanations in chain-of-thought prompting](https://md.research.elamur.ai/free/en/papers/turpin-2023-language-models-don-t-always/). In Proceedings of the 37th International Conference on Neural Information Processing Systems, NIPS ’23, Red Hook, NY, USA. C…
24. Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc Le, Ed Chi, Sharan Narang, Aakanksha Chowdhery, and Denny Zhou. 2022. [Self-consistency improves chain of thought reasoning in language models](https://md.research.elamur.ai/free/en/papers/wang-2022-self-consistency-improves-chain-thought/). arXiv preprint.
25. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed H. Chi, Quoc V. Le, and Denny Zhou. 2022. [Chain-of-thought prompting elicits reasoning in large language models](https://md.research.elamur.ai/free/en/papers/wei-2022-chain-thought-prompting-elicits-reasoning/). In Proceedings of the 36th International Conference on Neural Information Processing Systems, NIPS ’22, R…
26. Sarah Wiegreffe and Ana Marasović. 2021. Teach me to explain: A review of datasets for explainable natural language processing. arXiv preprint. [find](https://scholar.google.com/scholar?q=Sarah+Wiegreffe+and+Ana+Marasovic%CC%81%2E+2021%2E+Teach+me+to+explain%3A+A+review+of+datasets+for+explainable+natural+language+processing%2E+arXiv+preprint%2E)
27. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L. Griffiths, Yuan Cao, and Karthik Narasimhan. 2023. Tree of thoughts: deliberate problem solving with large language models. In Proceedings of the 37th International Conference on Neural Information Processing Systems, NIPS ’23, Red Hook, NY… [find](https://scholar.google.com/scholar?q=Shunyu+Yao%2C+Dian+Yu%2C+Jeffrey+Zhao%2C+Izhak+Shafran%2C+Thomas+L%2E+Griffiths%2C+Yuan+Cao%2C+and+Karthik+Narasimhan%2E+2023%2E+Tree+of+thoughts%3A+deliberate+problem+solving+with+large+language+models%2E+In+Proceedings+o)

*Updated 31 October 2026*