Elamur Research⌕
☰
home / papers / 2026 / Divergent Thinking: Escape the Homogeneity Trap in Generative Commonsense Reasoning

Divergent Thinking: Escape the Homogeneity Trap in Generative Commonsense Reasoning

Paper · 2026 · Yiheng Tao, Kaiwen Cheng, Zhiwei Nie et al.

Summary

The paper studies generative commonsense reasoning (GCR), where a model must write a coherent sentence that uses a given set of concepts and still makes commonsense sense. The authors find that standard ensemble methods such as Self-Consistency and Multi-Chain Reasoning get worse when they sample more chains from one strong LLM. They call this the "homogeneity trap": the chains cluster in a narrow semantic region, so they repeat the same biases instead of adding new evidence. The authors show that deep semantic diversity, not lexical variety, predicts how well candidates can be combined. Based on this, they propose Explore-then-Integrate. A high-entropy explorer (e.g., Qwen3-32B) produces varied candidates. A reward-model filter removes invalid ones. A strong integrator (e.g., GPT-4o) then merges fragments from several candidates into one output. A distilled student model can take over the integrator's role. A new provenance-aware evaluation suite checks that outputs combine fragments rather than just selecting the best candidate. The work matters because it shows that the source of diversity, not the number of samples, drives ensemble gains.

Key points

  • Main results: The method reaches 90.2% accuracy on NoRa (state of the art) and improves SPICE on CommonGen-Lite by more than 10%.
  • Filtering step: The Pretrained Reward Model scores each candidate on four axes: coherence, length, lexical coverage and information density. The top-K candidates go to the integrator.
  • Provenance-aware evaluation, heterogeneous vs. homogeneous inputs:
    • Source Entropy rises from 0.36 to 0.58.
    • MSIR (granularity of fusion) rises from 32.2% to 50.5%.
    • Conflict density among selected fragments falls from 0.49 to 0.38.
  • Distilled Integration: The compact student keeps almost the same fusion behaviour (Source Entropy 0.57, MSIR 49.5%). It also generalizes across datasets to DimonGen.
  • Limitation: Integration happens in a single pass. The authors list iterative variants, such as self-refinement and multi-turn interaction, as future work.

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

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

🧭 Topics 1Research subjects and languages the work is about.Reasoning · 1,426 works

Cite

Cite as Yiheng Tao, Kaiwen Cheng, Zhiwei Nie, Chang Liu and Jie Chen. 2026. Divergent Thinking: Escape the Homogeneity Trap in Generative Commonsense Reasoning. Findings of the Association for Computational Linguistics: ACL 2026.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.915

@inproceedings{tao2026divergent,
  title = {{Divergent Thinking: Escape the Homogeneity Trap in Generative Commonsense Reasoning}},
  author = {Yiheng Tao, Kaiwen Cheng, Zhiwei Nie, Chang Liu and Jie Chen},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.915},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.915}
}

References

55 references, 22 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Peter Anderson, Basura Fernando, Mark Johnson, and Stephen Gould. 2016. Spice: Semantic propositional image caption evaluation. In Computer Vision – ECCV 2016, pages 382–398.
  2. Jinze Bai, Shuai Bai, Yunfei Chu, Zeyu Cui, Kai Dang, Xiaodong Deng, Yang Fan, Wenbin Ge, Yu Han, Fei Huang, and 1 others. 2023. Qwen technical report. arXiv preprint arXiv:2309.16609.
  3. Satanjeev Banerjee and Alon Lavie. 2005. METEOR: An automatic metric for MT evaluation with improved correlation with human judgments. In Proceedings of the ACL Workshop on Intrinsic and Extrinsic Evaluation Measures for Machine Translation and/or Summarization, pages 65–72, Ann Arbor, Michigan. Ass…
  4. Zhijun Chen, Jingzheng Li, Pengpeng Chen, Zhuoran Li, Kai Sun, Yuankai Luo, Qianren Mao, Ming Li, Likang Xiao, Dingqi Yang, and 1 others. 2025. Harnessing multiple large language models: A survey on llm ensemble. arXiv preprint arXiv:2502.18036. — [arXiv:2502.18036](https://arxiv.org/abs/2502.18036) find
  5. Samuel Rhys Cox, Yunlong Wang, Ashraf Abdul, Christian von der Weth, and Brian Y. Lim. 2021. Directed diversity: Leveraging language embedding distances for collective creativity in crowd ideation. In Proceedings of the 2021 CHI Conference on Human Factors in Computing Systems, CHI ’21, New York, NY… find
  6. Wanqing Cui, Keping Bi, Jiafeng Guo, and Xueqi Cheng. 2024. MORE: Multi-mOdal REtrieval augmented generative commonsense reasoning. In Findings of the Association for Computational Linguistics: ACL 2024, pages 1178–1192, Bangkok, Thailand. Association for Computational Linguistics. find
  7. DeepSeek-AI. 2024. Deepseek-v3 technical report. Preprint. find
  8. Hanze Dong, Wei Xiong, Deepanshu Goyal, Rui Pan, Shizhe Diao, Jipeng Zhang, Kashun Shum, and Tong Zhang. 2023. Raft: Reward ranked finetuning for generative foundation model alignment. arXiv preprint arXiv:2304.06767.
  9. Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kulkarni, Tiago Ramalho, and 1 others. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783.
  10. Sebastian Farquhar, Jannik Kossen, Lorenz Kuhn, and Yarin Gal. 2024. Detecting hallucinations in large language models using semantic entropy. Nature, 630(8017):625–630. find
  11. Dan Friedman and Adji Bousso Dieng. 2023. The vendi score: A diversity evaluation metric for machine learning. Transactions on Machine Learning Research. find
  12. Jonathan Gordon and Benjamin Van Durme. 2013. Reporting bias and knowledge acquisition. In Proceedings of the 2013 workshop on Automated knowledge base construction, pages 25–30. find
  13. Jian Guan, Fei Huang, Zhihao Zhao, Xiaoyan Zhu, and Minlie Huang. 2020. A knowledge-enhanced pre-training model for commonsense story generation. Transactions of the Association for Computational Linguistics, page 93–108. find
  14. Neel Guha, Mayee Chen, Trevor Chow, Ishan Khare, and Christopher Re. 2024. Smoothie: Label free language model routing. Advances in Neural Information Processing Systems, 37:127645–127672. find
  15. Catherine Havasi, Robert Speer, and Jason Alonso. 2007. Conceptnet 3: a flexible, multilingual semantic network for common sense knowledge. In Recent advances in natural language processing, pages 27–29. John Benjamins Philadelphia, PA. find
  16. Xingwei He, Yeyun Gong, A-Long Jin, Weizhen Qi, Hang Zhang, Jian Jiao, Bartuer Zhou, Biao Cheng, Sm Yiu, and Nan Duan. 2022. Metric-guided distillation: Distilling knowledge from the metric to ranker and retriever for generative commonsense reasoning. In Proceedings of the 2022 Conference on Empiric… find
  17. Cheng-Yu Hsieh, Chun-Liang Li, Chih-Kuan Yeh, Hsiang-Fu Nakayama, James Lee, and Ankur Taly. 2023. Distilling step-by-step! outperforming larger language models with less training data and smaller model sizes. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics…
  18. Tiansheng Huang, Sihao Hu, Fatih Ilhan, Selim Furkan Tekin, Zachary Yahn, Yichang Xu, and Ling Liu. 2025. Safety tax: Safety alignment makes your large reasoning models less reasonable. arXiv preprint arXiv:2503.00555. — [arXiv:2503.00555](https://arxiv.org/abs/2503.00555) find
  19. Dongfu Jiang, Xiang Ren, and Bill Yuchen Lin. 2023. LLM-blender: Ensembling large language models with pairwise ranking and generative fusion. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 14165–14178, Toronto, Canada. Assoc…
  20. Liwei Jiang, Yuanjun Chai, Margaret Li, Mickel Liu, Raymond Fok, Nouha Dziri, Yulia Tsvetkov, Maarten Sap, and Yejin Choi. 2025. Artificial hivemind: The open-ended homogeneity of language models (and beyond). In The Thirty-ninth Annual Conference on Neural Information Processing Systems Datasets an… find
  21. Takeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo, and Yusuke Iwasawa. 2022. Large language models are zero-shot reasoners. In Advances in Neural Information Processing Systems, volume 35, pages 22199–22213.
  22. Haonan Li, Yeyun Gong, Jian Jiao, Ruofei Zhang, Timothy Baldwin, and Nan Duan. 2021. KFCNet: Knowledge filtering and contrastive learning for generative commonsense reasoning. In Findings of the Association for Computational Linguistics: EMNLP 2021, pages 2918–2928, Punta Cana, Dominican Republic. A… find
  23. Jiwei Li, Michel Galley, Chris Brockett, Jianfeng Gao, and Bill Dolan. 2016. A diversity-promoting objective function for neural conversation models. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pag…
  24. Junyou Li, Qin Zhang, Yangbin Yu, Qiang Fu, and Deheng Ye. 2024. More agents is all you need. arXiv preprint arXiv:2402.05120. find
  25. Xiaomin Li, Zhou Yu, Zhiwei Zhang, Xupeng Chen, Ziji Zhang, Yingying Zhuang, Narayanan Sadagopan, and Anurag Beniwal. 2025. When thinking fails: The pitfalls of reasoning for instruction-following in LLMs. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. find
  26. Bill Yuchen Lin, Wangchunshu Zhou, Ming Shen, Pei Zhou, Chandra Bhagavatula, Yejin Choi, and Xiang Ren. 2020. CommonGen: A constrained text generation challenge for generative commonsense reasoning. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 1823–1840, Online. As… find
  27. Teaching small language models to reason. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1773–1796. Also arXiv:2212.08410 (2022).
  28. Chin-Yew Lin. 2004. ROUGE: A package for automatic evaluation of summaries. In Text Summarization Branches Out, pages 74–81, Barcelona, Spain. Association for Computational Linguistics.
  29. OpenAI. 2023. Gpt-4 technical report. arXiv preprint arXiv:2303.08774.
  30. Yong Lin, Hangyu Lin, Wei Xiong, Shizhe Diao, Jianmeng Liu, Jipeng Zhang, Rui Pan, Haoxiang Wang, Wenbin Hu, Hanning Zhang, Hanze Dong, Renjie Pi, Han Zhao, Nan Jiang, Heng Ji, Yuan Yao, and Tong Zhang. 2024. Mitigating the alignment tax of RLHF. In Proceedings of the 2024 Conference on Empirical Me…
  31. Chenzhengyi Liu, Jie Huang, Kerui Zhu, and Kevin Chen-Chuan Chang. 2023. DimonGen: Diversified generative commonsense reasoning for explaining concept relationships. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 4719– 4731,… find
  32. Xin Liu, Dayiheng Liu, Baosong Yang, Haibo Zhang, Junwei Ding, Wenqing Yao, Weihua Luo, Haiying Zhang, and Jinsong Su. 2022. Kgr4: Retrieval, retrospect, refine and rethink for commonsense generation. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 36, pages 11029–11037. find
  33. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational…
  34. Amey P Pasarkar and Adji Bousso Dieng. 2024. Cousins of the vendi score: A family of similarity-based diversity metrics for science and machine learning. In International Conference on Artificial Intelligence and Statistics, pages 3808–3816. PMLR. find
  35. Nils Reimers and Iryna Gurevych. 2019. Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 3982–3992.
  36. Jacob Russin, Sam Whitman McGrath, and Danielle J. Williams. 2025. From frege to chatgpt: Compositionality in language, cognition, and deep neural networks. Preprint, arXiv:2405.15164. — [arXiv:2405.15164](https://arxiv.org/abs/2405.15164) find
  37. Ye Liu, Yao Wan, Lifang He, Hao Peng, and S Yu Philip. 2021. KG-BART: Knowledge graph-augmented BART for generative commonsense reasoning. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 35, pages 8411–8419. find
  38. Sania Sinha, Tanawan Premsri, and Parisa Kordjamshidi. 2024. A survey on compositional learning of AI models: Theoretical and experimental practices. Transactions on Machine Learning Research. Survey Certification. find
  39. Jinliang Lu, Ziliang Pang, Min Xiao, Yaochen Zhu, Rui Xia, and Jiajun Zhang. 2024. Merge, ensemble, and cooperate! a survey on collaborative strategies in the era of large language models. arXiv preprint arXiv:2407.06089. — [arXiv:2407.06089](https://arxiv.org/abs/2407.06089) find
  40. Xuezhi Wang, Jason Wei, Dale Schuurmans, Quoc Le, Ed Chi, and Denny Zhou. 2023. Self-consistency improves chain of thought reasoning in language models. In The Eleventh International Conference on Learning Representations.
  41. Bo Lv, Chen Tang, Yanan Zhang, Xin Liu, Ping Luo, and Yue Yu. 2024. URG: A unified ranking and generation method for ensembling language models. In Findings of the Association for Computational Linguistics: ACL 2024, pages 4421–4434, Bangkok, Thailand. Association for Computational Linguistics. find
  42. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Ed Chi, Quoc Le, and Denny Zhou. 2022. Chain of thought prompting elicits reasoning in large language models. In Advances in Neural Information Processing Systems, volume 35, pages 24824–24837.
  43. Aman Madaan, Niket Tandon, Prakhar Gupta, Skyler Hallinan, Luyu Gao, Sarah Wiegreffe, Uri Alon, Nouha Dziri, Shrimai Prabhumoye, Yiming Yang, and 1 others. 2023. Self-refine: Iterative refinement with self-feedback. In Advances in Neural Information Processing Systems, volume 36, pages 46534– 46594. find
  44. Lucie Charlotte Magister, Jonathan Mallinson, Jakub Adamek, Eric Malmi, and Aliaksei Severyn. 2023. find
  45. Fan Wu, Emily Black, and Varun Chandrasekaran. 2025. Generative monoculture in large language models. In The Thirteenth International Conference on Learning Representations. find
  46. Wei Xiong, Hanze Dong, Chenlu Ye, Ziqi Wang, Han Zhong, Heng Ji, Nan Jiang, and Tong Zhang. 2024. Iterative preference learning from human feedback: Bridging theory and practice for rlhf under kl-constraint. Preprint, arXiv:2312.11456. find
  47. Zhangyue Yin, Qiushi Sun, Cheng Chang, Qipeng Guo, Junqi Dai, Xuanjing Huang, and Xipeng Qiu. 2023. Exchange-of-thought: Enhancing large language model capabilities through cross-model communication. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 151… find
  48. Ori Yoran, Tomer Wolfson, Ben Bogin, Uri Katz, Daniel Deutch, and Jonathan Berant. 2023. Answering questions by meta-reasoning over multiple chains of thought. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 5942–5966, Singapore. Association for Compu…
  49. Wenhao Yu, Chenguang Zhu, Lianhui Qin, Zhihan Zhang, Tong Zhao, and Meng Jiang. 2022. Diversifying content generation for commonsense reasoning with mixture of knowledge graph experts. In Proceedings of the 2nd Workshop on Deep Learning on Graphs for Natural Language Processing (DLG4NLP 2022), pages… find
  50. Tianhui Zhang, Bei Peng, and Danushka Bollegala. 2025. Evaluating the evaluation of diversity in commonsense generation. In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 24258– 24275, Vienna, Austria. Association for Computatio… find
  51. Yue Zhang, Yafu Li, Leyang Cui, Deng Cai, Lemao Liu, Tingchen Fu, Xinting Huang, Enbo Zhao, Yu Zhang, Chen Xu, Yulong Chen, Longyue Wang, Anh Tuan Luu, Wei Bi, Freda Shi, and Shuming Shi. 2023a. Siren’s song in the ai ocean: A survey on hallucination in large language models. arXiv preprint arXiv:23…
  52. Zhuosheng Zhang, Aston Zhang, Mu Li, Hai Zhao, George Karypis, and Alex Smola. 2023b. Automatic chain of thought prompting in large language models. In The Eleventh International Conference on Learning Representations.
  53. Denny Zhou, Dale Schuurmans, Xuezhi Wang, Jason Huang, Jason Wei, Ed Chi, and Quoc Le. 2023. Least-to-most prompting enables complex reasoning in large language models. In The Eleventh International Conference on Learning Representations.
  54. Zhanke Zhou, Rong Tao, Jianing Zhu, Yiwen Luo, Zengmao Wang, and Bo Han. 2024. Can language models perform robust reasoning in chain-of-thought prompting with noisy rationales? In The Thirty-eighth Annual Conference on Neural Information Processing Systems. find
  55. Yaoming Zhu, Sidi Lu, Lei Zheng, Jiaxian Guo, Weinan Zhang, Jun Wang, and Yong Yu. 2018. Texygen: A benchmarking platform for text generation models. In The 41st International ACM SIGIR Conference on Research & Development in Information Retrieval, SIGIR ’18, page 1097–1100, New York, NY, USA. Assoc…

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/tao-2026-divergent-thinking-escape-homogeneity-trap/