TopoRefine: Iterative Refinement with Reasoning Topology as High-Level Feedback
Abstract
By leveraging effective signals to refine their outputs, large language models (LLMs) can achieve superior performance compared to single-pass outputs. However, internal signals often suffer from accumulated hallucinations and a lack of confidence, while external signals are typically difficult to obtain and apply, hindering the development of the self-refinement paradigm. In this work, we propose a novel method named TopoRefine, which integrates the reasoning topology within the model’s outputs as reliable high-level feedback. Specifically, TopoRefine encourages LLMs to explore reasoning paths without assuming a predefined direction for improvement, while using a consistency mechanism to prevent performance degradation. Reasoning topology can provide higher-level information as feedback, offering more precise semantics and facilitating self-analysis. Experiments on mathematical datasets and with various LLMs demonstrate significant improvements using our method. We also provide a detailed analysis of TopoRefine’s efficiency.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Cites
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372 · see citing works
LLaMA: Open and Efficient Foundation Language Models
Touvron, H., Lavril, T., Izacard, G., Martinet, X., Lachaux, M.-A., Lacroix, T., Rozière, B., Goyal, N., Hambro, E., Azhar, F., et al. (2023) · arXiv · doi:10.48550/arxiv.2302.13971 · Cited by 4,602 · see citing works
Chain-Of-Thought Prompting Elicits Reasoning in Large Language Models
Wei, J., Wang, X., Schuurmans, D., Bosma, M., Ichter, B., Xia, F., Chi, E., Le, Q. V., & Zhou, D. (2022) · Advances in Neural Information Processing Systems 35 · doi:10.52202/068431-1800 · Cited by 4,505 · see citing works
Training Language Models to Follow Instructions with Human Feedback
Ouyang, L., Wu, J., Jiang, X., Almeida, D., Wainwright, C., Mishkin, P., Zhang, C., Agarwal, S., Slama, K., Ray, A., et al. (2022) · Advances in Neural Information Processing Systems 35 · doi:10.52202/068431-2011 · Cited by 3,117 · see citing works
Emergent Abilities of Large Language Models
Wei, J., Tay, Y., Bommasani, R., Raffel, C., Zoph, B., Borgeaud, S., Yogatama, D., Bosma, M., Zhou, D., Metzler, D., et al. (2022) · arXiv · doi:10.48550/arxiv.2206.07682 · Cited by 1,264 · see citing works
Tree of Thoughts: Deliberate Problem Solving with Large Language Models
Yao, S., Yu, D., Zhao, J., Shafran, I., Griffiths, T. L., Cao, Y., & Narasimhan, K. (2023) · arXiv · doi:10.48550/arxiv.2305.10601 · Cited by 1,158 · see citing works
Reflexion: Language Agents with Verbal Reinforcement Learning
Shinn, N., Cassano, F., Berman, E., Gopinath, A., Narasimhan, K., & Yao, S. (2023) · arXiv · doi:10.48550/arxiv.2303.11366 · Cited by 1,087 · see citing works
Self-Refine: Iterative Refinement with Self-Feedback
Madaan, A., Tandon, N., Gupta, P., Hallinan, S., Gao, L., Wiegreffe, S., Alon, U., Dziri, N., Prabhumoye, S., Yang, Y., et al. (2023) · Advances in Neural Information Processing Systems 36 · doi:10.52202/075280-2019 · Cited by 755 · see citing works
Training Verifiers to Solve Math Word Problems
Cobbe, K., Kosaraju, V., Bavarian, M., Chen, M., Jun, H., Kaiser, L., Plappert, M., Tworek, J., Hilton, J., Nakano, R., et al. (2021) · arXiv · doi:10.48550/arxiv.2110.14168 · Cited by 575 · see citing works
Program Induction by Rationale Generation: Learning to Solve and Explain Algebraic Word Problems
Ling, W., Yogatama, D., Dyer, C., & Blunsom, P. (2017) · Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/p17-1015 · Cited by 325 · see citing works
Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate
Liang, T., He, Z., Jiao, W., Wang, X., Wang, Y., Wang, R., Yang, Y., Shi, S., & Tu, Z. (2023) · arXiv · doi:10.48550/arxiv.2305.19118 · Cited by 320 · see citing works
Are NLP Models really able to Solve Simple Math Word Problems?
Patel, A., Bhattamishra, S., & Goyal, N. (2021) · Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies · doi:10.18653/v1/2021.naacl-main.168 · Cited by 307 · see citing works
Encouraging Divergent Thinking in Large Language Models through Multi-Agent Debate
Liang, T., He, Z., Jiao, W., Wang, X., Wang, Y., Wang, R., Yang, Y., Shi, S., & Tu, Z. (2024) · Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing · doi:10.18653/v1/2024.emnlp-main.992 · Cited by 274 · see citing works
MathPrompter: Mathematical Reasoning using Large Language Models
Imani, S., Du, L., & Shrivastava, H. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-industry.4 · Cited by 212 · see citing works
Contrastive Decoding: Open-ended Text Generation as Optimization
Li, X. L., Holtzman, A., Fried, D., Liang, P., Eisner, J. M., Hashimoto, T., Zettlemoyer, L., & Lewis, M. (2023) · Proceedings of the conference - Association for Computational Linguistics. Meeting · doi:10.18653/v1/2023.acl-long.687 · Cited by 179 · see citing works
Progressive-Hint Prompting Improves Reasoning in Large Language ModelsPreprint
Zheng, C., Liu, Z.-Y., Xie, E.-Z., Li, Z.-G., & Li, Y. (2023) · arXiv.org · doi:10.48550/arxiv.2304.09797 · Cited by 165 · see citing works
RARR: Researching and revising what language models say, using language models
Gao, L., Dai, Z., Pasupat, P., Chen, A., Chaganty, A. T., Fan, Y., Zhao, V., Lao, N., Lee, H., Juan, D.-C., et al. (2023) · Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2023.acl-long.910 · Cited by 141 · see citing works
Bian, N., Han, X., Sun, L., Lin, H., Lu, Y., He, B., Jiang, S., & Dong, B. (2024) · Language Resources and Evaluation Conference (LREC) · doi:10.63317/32y85i5g9gso · Cited by 98 · see citing works
Lightman, H., Kosaraju, V., Burda, Y., Edwards, H., Baker, B., Lee, T., Leike, J., Schulman, J., Sutskever, I., & Cobbe, K. (2023) · arXiv · doi:10.48550/arxiv.2305.20050 · Cited by 65 · see citing works
ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs
Chen, J., Saha, S., & Bansal, M. (2024) · Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2024.acl-long.381 · Cited by 64 · see citing works
ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs
Chen, J. C.-Y., Saha, S., & Bansal, M. (2023) · arXiv · doi:10.48550/arxiv.2309.13007 · Cited by 64 · see citing works
Self-critiquing models for assisting human evaluators
Saunders, W., Yeh, C., Wu, J., Bills, S., Ouyang, L., Ward, J., & Leike, J. (2022) · arXiv · doi:10.48550/arxiv.2206.05802 · Cited by 50 · see citing works
Better Zero-Shot Reasoning with Self-Adaptive Prompting
Wan, X., Sun, R., Dai, H., Arık, S. Ö., & Pfister, T. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.216 · Cited by 35 · see citing works
LLMs cannot find reasoning errors, but can correct them given the error location
Tyen, G., Mansoor, H., Cărbune, V., Chen, P., & Mak, T. (2023) · arXiv · doi:10.48550/arxiv.2311.08516 · Cited by 32 · see citing works
Self-Contrast: Better Reflection Through Inconsistent Solving Perspectives
Zhang, W., Shen, Y., Wu, L., Peng, Q., Wang, J., Zhuang, Y., & Lu, W. (2024) · Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers) · doi:10.18653/v1/2024.acl-long.197 · Cited by 27 · see citing works
SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoning
Miao, N., Teh, Y. W., & Rainforth, T. (2023) · arXiv · doi:10.48550/arxiv.2308.00436 · Cited by 18 · cited 1× · see citing works
“Internal signals involve the model performing self-critique [25]–[27] or engaging in debates [28], [29] , which yield rich and comprehensive feedback.”
RRHF: Rank Responses to Align Language Models with Human Feedback
Yuan, H., Yuan, Z., Tan, C., Wang, W., Huang, S., & Huang, F. (2023) · neural information processing systems · doi:10.52202/075280-0482 · Cited by 17 · see citing works
Zeng, F., & Gao, W. G. (2023) · Findings of the Association for Computational Linguistics: ACL (Findings of ACL) · doi:10.18653/v1/2023.findings-acl.278 · Cited by 14 · see citing works
Self-Para-Consistency: Improving Reasoning Tasks at Low Cost for Large Language Models
Chen, W., Wang, W., Chu, Z., et al. (2024) · doi:10.18653/v1/2024.findings-acl.842 · Cited by 5 · see citing works
Confidence Matters: Revisiting Intrinsic Self-Correction Capabilities of Large Language ModelsPreprint
Li, L., Chen, Z., Chen, G., et al. (2024) · arXiv (Cornell University) · doi:10.48550/arxiv.2402.12563 · Cited by 4 · see citing works
Flagged first, then by citations
6 of the 36 references couldn’t be matched to a known paper.
Research works printed but not linked: 7
- Zhou, Least-to-most prompting enables complex reasoning in large language models
- Creswell, Selection-inference: Exploiting large language models for interpretable logical reasoning
- Wang, Self-consistency improves chain of thought reasoning in language models
- Gou, CRITIC: large language models can self-correct with tool-interactive critiquing
- Ouyang, 2022, Training language models to follow instructions with human feedback, Advances in Neural Information Processing Systems, 35, 27 730
- Huang, Large language models cannot self-correct reasoning yet
- Chen, Boosting of thoughts: Trial-and-error problem solving with large language models
- Authors
- Liao, H., Hu, S., Zhu, Z., He, H., & Jin, Y.
- Affiliations
- Shanghai Jiao Tong University (CN)
- Venue
- ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
- Publisher
- IEEE
- Published
- 2025-04-06
- Pages
- 1-5
- ISSN
- 1520-6149 (ISSN-L)
- OpenAlex
- W4408354006
- Semantic Scholar
- bf664a07197259d9aecf63cb52b2a313ac133104
- Cited by
- 0
- References
- 36
- FWCI
- 0.00
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Conference on Acoustics Speech and Signal Processing (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Multimodal Machine Learning Applications
- Keywords
- self-refinement, large language models, iterative refinement, hallucination mitigation, mathematical reasoning, self-analysis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
