---
title: "Where Reasoning Breaks — paper by Park and Lei (2026)"
description: "Building on this, they propose a framework that acts only at these junctions, at three levels… 38 references, cited by 1 paper here."
authors: ["Seunghyun Park", "Yuanyuan Lei"]
year: 2026
venue: "Findings of the Association for Computational Linguistics: ACL 2026"
source: "https://aclanthology.org/2026.findings-acl.1311/"
doi: "10.18653/v1/2026.findings-acl.1311"
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/park-2026-where-reasoning-breaks-logic-aware/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/park-2026-where-reasoning-breaks-logic-aware/)

[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/19/) / Where Reasoning Breaks: Logic-Aware Path Selection by Controlling Logical Connectives in LLMs Reasoning Chains

# Where Reasoning Breaks: Logic-Aware Path Selection by Controlling Logical Connectives in LLMs Reasoning Chains

**Authors:** [Seunghyun Park](https://md.research.elamur.ai/free/en/authors/seunghyun-park/) and [Yuanyuan Lei](https://md.research.elamur.ai/free/en/authors/yuanyuan-lei/) · **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.1311/) · [PDF](https://aclanthology.org/2026.findings-acl.1311.pdf) · [DOI 10.18653/v1/2026.findings-acl.1311](https://doi.org/10.18653/v1/2026.findings-acl.1311) · [OpenAlex](https://openalex.org/W7166823153)

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

## Summary
LLMs often fail at multi-step deduction because one wrong transition can derail the rest of the chain. The authors locate much of this fragility in explicit logical connectives such as "therefore", "but" and "however". Their diagnostic study, mainly on ZebraLogic with a BBH deductive subset for token-swap tests, finds three things. Model uncertainty clusters at connective positions. Competing top candidates mostly differ in which connective comes next. Swapping only the connective can flip the final answer.

Building on this, they propose a framework that acts only at these junctions, at three levels:
- **Gradient-based Logical Steering** adjusts activations without changing weights.
- **Localized Branching** runs a short look-ahead search when ambiguity at a connective is high.
- **Targeted Transition Preference Optimization (TTPO)** is a DPO-style objective applied to the single connective token.

Across five logical reasoning benchmarks and four LLMs, the methods improve accuracy while costing close to greedy decoding. Global scaling methods such as self-consistency and beam search need far more compute. The work suggests that small, linguistically grounded token-level interventions can complement or replace expensive global test-time scaling.

## Key points
- Changing a connective breaks correct reasoning chains roughly twice as often as changing other high-entropy tokens. Here, "high-entropy" means entropy above a threshold of τ = 1.0.
- With TTPO, Gemma-3-12b-it improves on ZebraLogic from 53.2 to 55.6.
- On Gemma-3-4b-it, the best method per task averages 65.7, compared with 65.2 for self-consistency (n=5).
  - It costs at most 1.45× greedy tokens, while self-consistency costs 2.90×.
  - On Phi-4-mini-instruct, self-consistency is slightly ahead (63.8 vs. 63.3) but uses 4.37× the tokens.
- The look-ahead score picks the ground-truth connective in 73.41% of cases (Gemma-3-4b-it) and 69.14% (Phi-4-mini-instruct).
- Gains depend on how many connectives a dataset contains. Improvements are smaller on BBH and LogiQA 2.0 than on ZebraLogic, and implicit reasoning steps without markers are not covered. Code is available at github.com/lei-nlp-lab/reasoning_fragility_acl_2026.

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

## Institutions

- [University of Florida](https://md.research.elamur.ai/free/en/universities/university-of-florida/) — University, Gainesville, United States · 33 works · 364 citations

📚 Shelves (4): [🖥️ 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

🧭 Topics (2): [🧠 Language models & pretraining](https://md.research.elamur.ai/free/en/topics/language-models/) · 3,320 works · [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,426 works

## Cite

Cite as: Seunghyun Park and Yuanyuan Lei. 2026. Where Reasoning Breaks: Logic-Aware Path Selection by Controlling Logical Connectives in LLMs Reasoning Chains. 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{park2026where,
  title = {{Where Reasoning Breaks: Logic-Aware Path Selection by Controlling Logical Connectives in LLMs Reasoning Chains}},
  author = {Seunghyun Park and Yuanyuan Lei},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.1311},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.1311}
}
```

- [← previous When Vision-Language Models Judge Without Seeing: Exposing Informativeness Bias](https://md.research.elamur.ai/free/en/papers/zou-2026-when-vision-language-models-judge/)
- [next → WhiteHouse: Translation of the Casablanca Corpus for Multi-dialectal Arabic Speech Translation](https://md.research.elamur.ai/free/en/papers/bougares-2026-whitehouse-translation-casablanca-corpus-multi/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/19/)

## References

38 references, 18 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Marah Abdin, Jyoti Aneja, Harkirat Behl, Sébastien Bubeck, Ronen Eldan, Suriya Gunasekar, Michael Harrison, Russell J Hewett, Mojan Javaheripi, Piero Kauffmann, and 1 others. 2024. [Phi-4 technical report](https://md.research.elamur.ai/free/en/papers/abdin-2024-phi-4-technical-report/). arXiv preprint arXiv:2412.08905.
2. Paul C Bogdan, Uzay Macar, Neel Nanda, and Arthur Conmy. 2025. Thought anchors: Which llm reasoning steps matter? arXiv preprint arXiv:2506.19143. [find](https://scholar.google.com/scholar?q=Paul+C+Bogdan%2C+Uzay+Macar%2C+Neel+Nanda%2C+and+Arthur+Conmy%2E+2025%2E+Thought+anchors%3A+Which+llm+reasoning+steps+matter%3F+arXiv+preprint+arXiv%3A2506%2E19143%2E)
3. Charlie Chen, Sebastian Borgeaud, Geoffrey Irving, Jean-Baptiste Lespiau, Laurent Sifre, and John Jumper. 2023. Accelerating large language model decoding with speculative sampling. arXiv preprint arXiv:2302.01318. [find](https://scholar.google.com/scholar?q=Charlie+Chen%2C+Sebastian+Borgeaud%2C+Geoffrey+Irving%2C+Jean%2DBaptiste+Lespiau%2C+Laurent+Sifre%2C+and+John+Jumper%2E+2023%2E+Accelerating+large+language+model+decoding+with+speculative+sampling%2E+arXiv+preprint+arX)
4. Qiguang Chen, Libo Qin, Jinhao Liu, Dengyun Peng, Jiannan Guan, Peng Wang, Mengkang Hu, Yuhang Zhou, Te Gao, and Wanxiang Che. 2025. [Towards reasoning era: A survey of long chain-of-thought for reasoning large language models](https://md.research.elamur.ai/free/en/papers/chen-2025-towards-reasoning-era-survey-long/). arXiv preprint arXiv:2503.09567.
5. Antonia Creswell and Murray Shanahan. 2022. Faithful reasoning using large language models. arXiv preprint arXiv:2208.14271. [find](https://scholar.google.com/scholar?q=Antonia+Creswell+and+Murray+Shanahan%2E+2022%2E+Faithful+reasoning+using+large+language+models%2E+arXiv+preprint+arXiv%3A2208%2E14271%2E)
6. Antonia Creswell, Murray Shanahan, and Irina Higgins. 2022. Selection-inference: Exploiting large language models for interpretable logical reasoning. arXiv preprint arXiv:2205.09712. [find](https://scholar.google.com/scholar?q=Antonia+Creswell%2C+Murray+Shanahan%2C+and+Irina+Higgins%2E+2022%2E+Selection%2Dinference%3A+Exploiting+large+language+models+for+interpretable+logical+reasoning%2E+arXiv+preprint+arXiv%3A2205%2E09712%2E)
7. Yichao Fu, Xuewei Wang, Yuandong Tian, and Jiawei Zhao. 2025. Deep think with confidence. arXiv preprint arXiv:2508.15260. [find](https://scholar.google.com/scholar?q=Yichao+Fu%2C+Xuewei+Wang%2C+Yuandong+Tian%2C+and+Jiawei+Zhao%2E+2025%2E+Deep+think+with+confidence%2E+arXiv+preprint+arXiv%3A2508%2E15260%2E)
8. Etash Guha, Ryan Marten, Sedrick Keh, Negin Raoof, Georgios Smyrnis, Hritik Bansal, Marianna Nezhurina, Jean Mercat, Trung Vu, Zayne Sprague, and 1 others. 2025. [Openthoughts: Data recipes for reasoning models](https://md.research.elamur.ai/free/en/papers/guha-2025-openthoughts-data-recipes-reasoning-models/). arXiv preprint arXiv:2506.04178.
9. Junxian He, Yu Wu, Junlong Li, Daya Guo, Dejian Yang, and Runxin Xu. 2025. [Codei/o: Condensing reasoning patterns via code input-output prediction](https://md.research.elamur.ai/free/en/papers/li-2025-codei-o-condensing-reasoning/).
10. Yuanyuan Lei and Ruihong Huang. 2024. Boosting logical fallacy reasoning in llms via logical structure tree. arXiv preprint arXiv:2410.12048. [find](https://scholar.google.com/scholar?q=Yuanyuan+Lei+and+Ruihong+Huang%2E+2024%2E+Boosting+logical+fallacy+reasoning+in+llms+via+logical+structure+tree%2E+arXiv+preprint+arXiv%3A2410%2E12048%2E)
11. Hunter Lightman, Vineet Kosaraju, Yuri Burda, Harrison Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe. 2023. [Let’s verify step by step](https://md.research.elamur.ai/free/en/papers/lightman-2023-let-s-verify-step-step/). In The Twelfth International Conference on Learning Representations.
12. Bill Yuchen Lin, Ronan Le Bras, Kyle Richardson, Ashish Sabharwal, Radha Poovendran, Peter Clark, and Yejin Choi. 2025. Zebralogic: On the scaling limits of llms for logical reasoning. arXiv preprint arXiv:2502.01100. [find](https://scholar.google.com/scholar?q=Bill+Yuchen+Lin%2C+Ronan+Le+Bras%2C+Kyle+Richardson%2C+Ashish+Sabharwal%2C+Radha+Poovendran%2C+Peter+Clark%2C+and+Yejin+Choi%2E+2025%2E+Zebralogic%3A+On+the+scaling+limits+of+llms+for+logical+reasoning%2E+arXiv+preprint+)
13. Zicheng Lin, Tian Liang, Jiahao Xu, Qiuzhi Lin, Xing Wang, Ruilin Luo, Chufan Shi, Siheng Li, Yujiu Yang, and Zhaopeng Tu. 2024. Critical tokens matter: Token-level contrastive estimation enhances llm’s reasoning capability. arXiv preprint arXiv:2411.19943. — [arXiv:2411.19943](https://arxiv.org/abs/2411.19943) [find](https://scholar.google.com/scholar?q=Zicheng+Lin%2C+Tian+Liang%2C+Jiahao+Xu%2C+Qiuzhi+Lin%2C+Xing+Wang%2C+Ruilin+Luo%2C+Chufan+Shi%2C+Siheng+Li%2C+Yujiu+Yang%2C+and+Zhaopeng+Tu%2E+2024%2E+Critical+tokens+matter%3A+Token%2Dlevel+contrastive+estimation+enhances+llm)
14. Chengwu Liu, Ye Yuan, Yichun Yin, Yan Xu, Xin Xu, Zaoyu Chen, Yasheng Wang, Lifeng Shang, Qun Liu, and Ming Zhang. 2025. Safe: Enhancing mathematical reasoning in large language models via retrospective step-aware formal verification. arXiv preprint arXiv:2506.04592. [find](https://scholar.google.com/scholar?q=Chengwu+Liu%2C+Ye+Yuan%2C+Yichun+Yin%2C+Yan+Xu%2C+Xin+Xu%2C+Zaoyu+Chen%2C+Yasheng+Wang%2C+Lifeng+Shang%2C+Qun+Liu%2C+and+Ming+Zhang%2E+2025%2E+Safe%3A+Enhancing+mathematical+reasoning+in+large+language+models+via+retrospecti)
15. Hanmeng Liu, Jian Liu, Leyang Cui, Zhiyang Teng, Nan Duan, Ming Zhou, and Yue Zhang. 2023. Logiqa 2.0—an improved dataset for logical reasoning in natural language understanding. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 31:2947–2962. [find](https://scholar.google.com/scholar?q=Hanmeng+Liu%2C+Jian+Liu%2C+Leyang+Cui%2C+Zhiyang+Teng%2C+Nan+Duan%2C+Ming+Zhou%2C+and+Yue+Zhang%2E+2023%2E+Logiqa+2%2E0%E2%80%94an+improved+dataset+for+logical+reasoning+in+natural+language+understanding%2E+IEEE%2FACM+Transactions)
16. Jiebo Luo, Xin Qian, Julian McAuley, Yuwei Cao, Daoan Zhang, Tianyang Liu, Xiaoyi Liu, Antoine Simoulin, Grey Yang, Dayu Yang, and Zhaopu Teng. 2025. [Code to think, think to code: A survey on code-enhanced reasoning and reasoning-driven code intelligence in llms](https://md.research.elamur.ai/free/en/papers/yang-2025-code-think-think-code-survey/).
17. Lachlan McGinness and Peter Baumgartner. 2024. Automated theorem provers help improve large language model reasoning. arXiv preprint arXiv:2408.03492. [find](https://scholar.google.com/scholar?q=Lachlan+McGinness+and+Peter+Baumgartner%2E+2024%2E+Automated+theorem+provers+help+improve+large+language+model+reasoning%2E+arXiv+preprint+arXiv%3A2408%2E03492%2E)
18. Liangming Pan, Alon Albalak, Xinyi Wang, and William Wang. 2023. [Logic-lm: Empowering large language models with symbolic solvers for faithful logical reasoning](https://md.research.elamur.ai/free/en/papers/pan-2023-logic-lm-empowering-large-language/). In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 3806–3824.
19. Nina Panickssery, Nick Gabrieli, Julian Schulz, Meg Tong, Evan Hubinger, and Alexander Matt Turner. 2024. Steering llama 2 via contrastive activation addition, 2024. 3. [find](https://scholar.google.com/scholar?q=Nina+Panickssery%2C+Nick+Gabrieli%2C+Julian+Schulz%2C+Meg+Tong%2C+Evan+Hubinger%2C+and+Alexander+Matt+Turner%2E+2024%2E+Steering+llama+2+via+contrastive+activation+addition%2C+2024%2E+3%2E)
20. Nina Rimsky, Nick Gabrieli, Julian Schulz, Meg Tong, Evan Hubinger, and Alexander Turner. 2024. Steering llama 2 via contrastive activation addition. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 15504–15522. [find](https://scholar.google.com/scholar?q=Nina+Rimsky%2C+Nick+Gabrieli%2C+Julian+Schulz%2C+Meg+Tong%2C+Evan+Hubinger%2C+and+Alexander+Turner%2E+2024%2E+Steering+llama+2+via+contrastive+activation+addition%2E+In+Proceedings+of+the+62nd+Annual+Meeting+of+the+A)
21. Livio Robaldo. 2008. [The penn discourse treebank 2.0](https://md.research.elamur.ai/free/en/papers/prasad-2008-penn-discourse-treebank-2-0/). In Lrec.
22. Mohammed Saeed, Naser Ahmadi, Preslav Nakov, and Paolo Papotti. 2021. Rulebert: Teaching soft rules to pre-trained language models. arXiv preprint arXiv:2109.13006. [find](https://scholar.google.com/scholar?q=Mohammed+Saeed%2C+Naser+Ahmadi%2C+Preslav+Nakov%2C+and+Paolo+Papotti%2E+2021%2E+Rulebert%3A+Teaching+soft+rules+to+pre%2Dtrained+language+models%2E+arXiv+preprint+arXiv%3A2109%2E13006%2E)
23. Abulhair Saparov and He He. 2022. [Language models are greedy reasoners: A systematic formal analysis of chain-of-thought](https://md.research.elamur.ai/free/en/papers/saparov-2022-language-models-greedy-reasoners-systematic/). arXiv preprint arXiv:2210.01240.
24. Mirac Suzgun, Nathan Scales, Nathanael Schärli, Sebastian Gehrmann, Yi Tay, Hyung Won Chung, Aakanksha Chowdhery, Quoc Le, Ed Chi, Denny Zhou, and 1 others. 2023. [Challenging big-bench tasks and whether chain-of-thought can solve them](https://md.research.elamur.ai/free/en/papers/suzgun-2023-challenging-big-bench-tasks-whether/). In Findings of the Association for Computational Linguistics: AC…
25. Gemma Team, Aishwarya Kamath, Johan Ferret, Shreya Pathak, Nino Vieillard, Ramona Merhej, Sarah Perrin, Tatiana Matejovicova, Alexandre Ramé, Morgane Rivière, and 1 others. 2025. Gemma 3 technical report. arXiv preprint arXiv:2503.19786. [find](https://scholar.google.com/scholar?q=Gemma+Team%2C+Aishwarya+Kamath%2C+Johan+Ferret%2C+Shreya+Pathak%2C+Nino+Vieillard%2C+Ramona+Merhej%2C+Sarah+Perrin%2C+Tatiana+Matejovicova%2C+Alexandre+Ram%C3%A9%2C+Morgane+Rivi%C3%A8re%2C+and+1+others%2E+2025%2E+Gemma+3+technical+rep)
26. Alexander Matt Turner, Lisa Thiergart, Gavin Leech, David Udell, Juan J Vazquez, Ulisse Mini, and Monte MacDiarmid. 2023. [Steering language models with activation engineering](https://md.research.elamur.ai/free/en/papers/turner-2023-steering-language-models-activation-engineering/), 2023.
27. Shenzhi Wang, Le Yu, Chang Gao, Chujie Zheng, Shixuan Liu, Rui Lu, Kai Dang, Xionghui Chen, Jianxin Yang, Zhenru Zhang, and 1 others. 2025. [Beyond the 80/20 rule: High-entropy minority tokens drive effective reinforcement learning for llm reasoning](https://md.research.elamur.ai/free/en/papers/wang-2025-80-20-rule-high-entropy/). arXiv preprint arXiv:2506.01939.
28. 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 arXiv:2203.11171.
29. Furu Wei, Wenshan Wu, Shaoguang Mao, Yadong Zhang, Yan Xia, Tao Ge, Man Lan, Xun Wang, Ting Song, and Adrian de Wynter. 2024. Llm as a mastermind: A survey of strategic reasoning with large language models. [find](https://scholar.google.com/scholar?q=Furu+Wei%2C+Wenshan+Wu%2C+Shaoguang+Mao%2C+Yadong+Zhang%2C+Yan+Xia%2C+Tao+Ge%2C+Man+Lan%2C+Xun+Wang%2C+Ting+Song%2C+and+Adrian+de+Wynter%2E+2024%2E+Llm+as+a+mastermind%3A+A+survey+of+strategic+reasoning+with+large+language+m)
30. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Ed H. Chi, F. Xia, Quoc Le, and Denny Zhou. 2022a. [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/). ArXiv, abs/2201.11903.
31. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, and 1 others. 2022b. [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/). Advances in neural information processing systems, 35:24824– 24837.
32. Zonglin Yang, Xinya Du, Rui Mao, Jinjie Ni, and Erik Cambria. 2023. [Logical reasoning over natural language as knowledge representation: A survey](https://md.research.elamur.ai/free/en/papers/yang-2023-logical-reasoning-over-natural-language/). arXiv preprint arXiv:2303.12023.
33. 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, 2023. 3:1. [find](https://scholar.google.com/scholar?q=Shunyu+Yao%2C+Dian+Yu%2C+Jeffrey+Zhao%2C+Izhak+Shafran%2C+Thomas+L+Griffiths%2C+Yuan+Cao%2C+and+Karthik+Narasimhan%2E+2023%2E+Tree+of+thoughts%3A+Deliberate+problem+solving+with+large+language+models%2C+2023%2E+3%3A1%2E)
34. Jiaxin Zhang, Zhijiang Guo, Haotian Xu, Yingying Zhang, Jiahua Dong, Yuxuan Yao, Le Song, Cheng-Lin Liu, Junhao Zheng, Fei Yin, Duzhen Zhang, Xiuyi Chen, Zhong-Zhi Li, Ming-Liang Zhang, Zengyan Liu, and Pei-Jie Wang. 2025a. [From system 1 to system 2: A survey of reasoning large language models](https://md.research.elamur.ai/free/en/papers/li-2025-system-1-system-2-survey/).
35. Kun Zhang, Zhouchen Lin, Haoxuan Li, Robert van Rooij, Fengxiang Cheng, and Fenrong Liu. 2025b. Empowering llms with logical reasoning: A comprehensive survey. [find](https://scholar.google.com/scholar?q=Kun+Zhang%2C+Zhouchen+Lin%2C+Haoxuan+Li%2C+Robert+van+Rooij%2C+Fengxiang+Cheng%2C+and+Fenrong+Liu%2E+2025b%2E+Empowering+llms+with+logical+reasoning%3A+A+comprehensive+survey%2E)
36. Yue Zhang, Chaoli Zhang, Hanmeng Liu, Ruoxi Ning, Mengru Ding, Zhizhang Fu, and Xiaozhang Liu. 2025c. Logical reasoning in [large language models: A survey](https://md.research.elamur.ai/free/en/papers/minaee-2024-large-language-models-survey/).
37. Yao Zhao, Zhitian Xie, Chen Liang, Chenyi Zhuang, and Jinjie Gu. 2024. Lookahead: An inference acceleration framework for large language model with lossless generation accuracy. In Proceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pages 6344–6355. [find](https://scholar.google.com/scholar?q=Yao+Zhao%2C+Zhitian+Xie%2C+Chen+Liang%2C+Chenyi+Zhuang%2C+and+Jinjie+Gu%2E+2024%2E+Lookahead%3A+An+inference+acceleration+framework+for+large+language+model+with+lossless+generation+accuracy%2E+In+Proceedings+of+the+)
38. Amir Zur, Atticus Geiger, Ekdeep Singh Lubana, and Eric Bigelow. 2025. Are language models aware of the road not taken? token-level uncertainty and hidden state dynamics. arXiv preprint arXiv:2511.04527. — [arXiv:2511.04527](https://arxiv.org/abs/2511.04527) [find](https://scholar.google.com/scholar?q=Amir+Zur%2C+Atticus+Geiger%2C+Ekdeep+Singh+Lubana%2C+and+Eric+Bigelow%2E+2025%2E+Are+language+models+aware+of+the+road+not+taken%3F+token%2Dlevel+uncertainty+and+hidden+state+dynamics%2E+arXiv+preprint+arXiv%3A2511%2E045)

*Updated 1 November 2026*