Elamur Research⌕
☰
home / papers / 2026 / Bridging Cognition and Affect: Emotion-Aware Opinion Summarization using LLMs

Bridging Cognition and Affect: Emotion-Aware Opinion Summarization using LLMs

Summary

Product review summarization systems usually reduce customer sentiment to polarity (positive, neutral, or negative). As a result, they do not explain which specific emotions lie behind the ratings. The authors propose a new task: emotion-aware opinion summarization (EAOS). It is based on the eight basic emotions from Plutchik's model. For generation, they built the EAOS-GEN framework. It is a single prompt that guides an LLM through several stages: link product aspects to emotions, balance majority and minority emotions, and then weave them into the text without explicit labels. For evaluation, they built the EAOS-EVAL framework. They also assembled the synthetic dataset EAOS-SUMM (40K "product — summary" pairs), the benchmark EAOS-EVAL-BENCH, and a distilled 1B-parameter model. It is not statistically inferior to Llama-3.1-70B. In a user study, 82% of readers preferred the emotion-aware summaries. The value of the work is that emotion analysis moves from classification to generating coherent text. In addition, the work shows that even a compact, deployable model can handle such a task.

Key points

  • The EAOS-SUMM dataset was generated by two teacher models, Phi-3.5-mini and Mistral-7B, and then spot-checked by humans. At the balancing stage, an emotion must be included in the summary if it occurs in at least 10% of reviews. Rarer emotions are discarded.
  • Evaluation covers 7 dimensions. Three of them are standard: FL, CO, FA. Four new ones concern emotions: EA, ESC, EBM, CER. Inter-annotator agreement is high (α = 0.85).
  • EAOS-Llama-3.2-1B has 70 times fewer parameters than Llama-3.1-70B and is 13% better than its base model without fine-tuning. The model is published on Hugging Face (cuteasduck/EAOS-Llama-3.2-1B), and the prompts on GitHub (arnavcse/EAOS).
  • Human experts scored 4.80 points on average. The best models reach 96% of this level. Emotion–aspect pairs from humans and LLMs overlap with a Jaccard score of 0.73. User study result: 82% (p < .001).
  • Limitations: the teacher models show a noticeable positivity bias. Joy and Trust are represented 4–6% more often than in the human annotation, and Anger and Disgust 2–4% less often.

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: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.findings-acl.221

@inproceedings{attri2026bridging,
  title = {{Bridging Cognition and Affect: Emotion-Aware Opinion Summarization using LLMs}},
  author = {Attri, Arnav and Attri, Anuj},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2026},
  year = {2026},
  doi = {10.18653/v1/2026.findings-acl.221},
  url = {https://doi.org/10.18653/v1/2026.findings-acl.221}
}

References

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

  1. AI@Meta. 2024. Llama 3 model card. find
  2. Hassan Alhuzali and Sophia Ananiadou. 2021. SpanEmo: Casting multi-label emotion classification as span-prediction. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 1573–1584, Online. Association for Computational Ling… find
  3. Reinald Kim Amplayo, Stefanos Angelidis, and Mirella Lapata. 2021. Aspect-controllable opinion summarization. arXiv preprint arXiv:2109.03171. find
  4. Reinald Kim Amplayo and Mirella Lapata. 2020. Unsupervised opinion summarization with noising and denoising. arXiv preprint arXiv:2004.10150. find
  5. Anuj Attri, Arnav Attri, Pushpak Bhattacharyya, Suman Banerjee, Amey Patil, Muthusamy Chelliah, and Nikesh Garera. 2025a. Llms as architects and critics for multi-source opinion summarization. find
  6. Arnav Attri, Anuj Attri, Suman Banerjee, Amey Patil, Muthusamy Chelliah, Nikesh Garera, and Pushpak Bhattacharyya. 2025b. Why we feel what we feel: Joint detection of emotions and their opinion triggers in E-commerce. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 55… find
  7. Manik Bhandari, Pranav Narayan Gour, Atabak Ashfaq, Pengfei Liu, and Graham Neubig. 2020. Reevaluating evaluation in text summarization. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 9347–9359, Online. Association for Computational Linguisti… find
  8. Adithya Bhaskar, Alex Fabbri, and Greg Durrett. 2023. Prompted opinion summarization with GPT-3.5. In Findings of the Association for Computational Linguistics: ACL 2023, pages 9282–9300, Toronto, Canada. Association for Computational Linguistics. find
  9. Laura Ana Maria Bostan, Evgeny Kim, and Roman Klinger. 2020. GoodNewsEveryone: A corpus of news headlines annotated with emotions, semantic roles, and reader perception. In Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 1554–1566, Marseille, France. European Language… find
  10. Arthur Bražinskas, Mirella Lapata, and Ivan Titov. 2020. Few-shot learning for opinion summarization. arXiv preprint arXiv:2004.14884. find
  11. Tao Chen, Premaratne Samaranayake, XiongYing Cen, Meng Qi, and Yi-Chen Lan. 2022. The impact of online reviews on consumers’ purchasing decisions: Evidence from an eye-tracking study. Frontiers in Psychology, 13:865702. find
  12. Cheng-Han Chiang and Hung-yi Lee. 2023a. Can large language models be an alternative to human evaluations? In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 15607–15631, Toronto, Canada. Association for Computational Linguistics… find
  13. Cheng-Han Chiang and Hung-yi Lee. 2023b. A closer look into using large language models for automatic evaluation. In Findings of the Association for Computational Linguistics: EMNLP 2023, pages 8928– 8942, Singapore. Association for Computational Linguistics. find
  14. Eric Chu and Peter Liu. 2019. Meansum: a neural model for unsupervised multi-document abstractive summarization. In International conference on machine learning, pages 1223–1232. PMLR. find
  15. Jonathan H. Clark, Chris Dyer, Alon Lavie, and Noah A. Smith. 2011. Better hypothesis testing for statistical machine translation: Controlling for optimizer instability. In Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, pages 176… find
  16. Jacob Cohen. 1988. Statistical Power Analysis for the Behavioral Sciences, 2nd edition. Lawrence Erlbaum Associates, Hillsdale, NJ. find
  17. Bjarke Felbo, Alan Mislove, Anders Søgaard, Iyad Rah-wan, and Sune Lehmann. 2017. Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 161… find our card
  18. Jinlan Fu, See-Kiong Ng, Zhengbao Jiang, and Pengfei Liu. 2023. Gptscore: Evaluate as you desire. find
  19. Dan Gillick and Yang Liu. 2010. Non-expert evaluation of summarization systems is risky. In Proceedings of the NAACL HLT 2010 Workshop on Creating Speech and Language Data with Amazon’s Mechanical Turk, pages 148–151, Los Angeles. Association for Computational Linguistics. find
  20. Junpeng Guo, Xiaopan Wang, and Yi Wu. 2020. Positive emotion bias: Role of emotional content from online customer reviews in purchase decisions. J. Retail. Consum. Serv., 52(101891):101891. find
  21. Rishav Hada, Varun Gumma, Mohamed Ahmed, Kalika Bali, and Sunayana Sitaram. 2024. METAL: Towards multilingual meta-evaluation. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 2280–2298, Mexico City, Mexico. Association for Computational Linguistics. find
  22. Weihua Deng, Ming Yi, and Yingying Lu. 2020. Vote or not? how various information cues affect helpfulness voting of online reviews. Online Inf. Rev., 44(4):787– 803. find
  23. Tom Hosking, Hao Tang, and Mirella Lapata. 2023. Attributable and scalable opinion summarization. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8488–8505, Toronto, Canada. Association for Computational Linguistics. find
  24. Aniket Deroy, Kripabandhu Ghosh, and Saptarshi Ghosh. 2023. How ready are pre-trained abstractive models and llms for legal case judgement summarization? find
  25. Yupeng Hou, Jiacheng Li, Zhankui He, An Yan, Xiusi Chen, and Julian McAuley. 2024. Bridging language and items for retrieval and recommendation. arXiv preprint arXiv:2403.03952. — [arXiv:2403.03952](https://arxiv.org/abs/2403.03952) find
  26. Zixiang Ding, Rui Xia, and Jianfei Yu. 2020. Ecpe-2d: Emotion-cause pair extraction based on joint two-dimensional representation, interaction and prediction. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 3161–3170. find
  27. Chi Hu, Yuan Ge, Xiangnan Ma, Hang Cao, Qiang Li, Yonghua Yang, Tong Xiao, and Jingbo Zhu. 2024. Rankprompt: Step-by-step comparisons make language models better reasoners. find
  28. Paul Ekman. 1992. An argument for basic emotions. Cogn. Emot., 6(3-4):169–200. find our card
  29. Günes Erkan and Dragomir R Radev. 2004. Lexrank: Graph-based lexical centrality as salience in text summarization. Journal of artificial intelligence research, 22:457–479. find
  30. Alexander R. Fabbri, Wojciech Kryściński, Bryan Mc-Cann, Caiming Xiong, Richard Socher, and Dragomir Radev. 2021. Summeval: Re-evaluating summarization evaluation. find
  31. Armin Felbermayr and Alexandros Nanopoulos. 2016. The role of emotions for the perceived usefulness in online customer reviews. Journal of Interactive Marketing, 36:60–76. find
  32. Albert Q. Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, Lélio Renard Lavaud, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix,… find
  33. Albert Q. Jiang, Alexandre Sablayrolles, Antoine Roux, Arthur Mensch, Blanche Savary, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Emma Bou Hanna, Florian Bressand, Gianna Lengyel, Guillaume Bour, Guillaume Lample, Lélio Renard Lavaud, Lucile Saulnier, Marie-Anne Lachaux, Pierre Stock,… find
  34. Hyun Duk Kim, Kavita Ganesan, Parikshit Sondhi, and ChengXiang Zhai. 2011. Comprehensive review of opinion summarization. find
  35. Tom Kocmi and Christian Federmann. 2023. Large language models are state-of-the-art evaluators of translation quality. In Proceedings of the 24th Annual Conference of the European Association for Machine Translation, pages 193–203, Tampere, Finland. European Association for Machine Translation. find
  36. Paulo Lopes, John Nezlek, Astrid Schuetz, Ina Sellin, and Peter Salovey. 2004. Emotional intelligence and social interaction. Personality social psychology bulletin, 30:1018–34. find
  37. M S I Malik and Ayyaz Hussain. 2017. Helpfulness of product reviews as a function of discrete positive and negative emotions. Comput. Human Behav., 73:290– 302. find
  38. Klaus Krippendorff. 2011. Computing krippendorff’s alpha-reliability. find
  39. Dina Mayzlin, Yaniv Dover, and Judith A Chevalier. 2012. Promotional reviews: An empirical investigation of online review manipulation. SSRN Electron. J. find
  40. Theodoros Lappas, Gaurav Sabnis, and Georgios Valka-nas. 2016. The impact of fake reviews on online visibility: A vulnerability assessment of the hotel industry. Inf. Syst. Res., 27(4):940–961. find
  41. Xiao-Li Meng, Robert Rosenthal, and Donald B. Rubin. 1992. Comparing correlated correlation coefficients. Psychological Bulletin, 111(1):172–175. find
  42. Mike Lewis, Yinhan Liu, Naman Goyal, Marjan Ghazvininejad, Abdel rahman Mohamed, Omer Levy, Veselin Stoyanov, and Luke Zettlemoyer. 2019. Bart: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. In Annual Meeting of the Association for Comput… find our card
  43. Haoyuan Li, Somnath Basu Roy Chowdhury, and Snigdha Chaturvedi. 2023. Aspect-aware unsupervised extractive opinion summarization. In Findings of the Association for Computational Linguistics: ACL 2023, pages 12662–12678, Toronto, Canada. Association for Computational Linguistics. find
  44. Haoyuan Li and Snigdha Chaturvedi. 2024. Rationale-based opinion summarization. In Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers), pages 8274–8292, Mexico City, Mexico. Association… find
  45. Jinfen Li and Lu Xiao. 2023. Multi-emotion recognition using Multi-EmoBERT and emotion analysis in fake news. In Proceedings of the 15th ACM Web Science Conference 2023, New York, NY, USA. ACM. find
  46. 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. find
  47. Yang Liu, Dan Iter, Yichong Xu, Shuohang Wang, Ruochen Xu, and Chenguang Zhu. 2023. G-eval: NLG evaluation using gpt-4 with better human alignment. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 2511–2522, Singapore. Association for Computational Lin… find
  48. Yixin Liu, Kejian Shi, Katherine S He, Longtian Ye, Alexander R. Fabbri, Pengfei Liu, Dragomir Radev, and Arman Cohan. 2024. On learning to summarize with large language models as references. find
  49. Saif Mohammad and Felipe Bravo-Marquez. 2017. Emotion intensities in tweets. In Proceedings of the 6th Joint Conference on Lexical and Computational Semantics (*SEM 2017), pages 65–77, Vancouver, Canada. Association for Computational Linguistics. find our card
  50. Saif M. Mohammad and Peter D. Turney. 2013. Crowdsourcing a word–emotion association lexicon. Computational Intelligence, 29(3):436–465. find
  51. OpenAI. 2023. Gpt-4 technical report. abs/2303.08774. find our card
  52. ArXiv, 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 Comput… find our card
  53. Ilias O Pappas, Panos E Kourouthanassis, Michail N Gi-annakos, and Vassilios Chrissikopoulos. 2014. Shiny happy people buying: the role of emotions on personalized e-shopping. Electron. Mark., 24(3):193–206. find
  54. Ethan Perez, Sam Ringer, Kamile Lukosiute, Karina Nguyen, Edwin Chen, Scott Heiner, Craig Pettit, Catherine Olsson, Sandipan Kundu, Saurav Kadavath, Andy Jones, Anna Chen, Benjamin Mann, Brian Israel, Bryan Seethor, Cameron McKinnon, Christopher Olah, Da Yan, Daniela Amodei, Dario Amodei, Dawn Drain… find our card
  55. Roberts, Aditya Barua, Alex Botev, Alex Castro-Ros, Ambrose Slone, Amélie Héliou, Andrea Tacchetti, Anna Bulanova, Antonia Paterson, Beth Tsai, Bobak Shahriari, Charline Le Lan, Christopher A. Choquette-Choo, Clément Crepy, Daniel Cer, Daphne Ippolito, David Reid, Elena Buchatskaya, Eric Ni, Eric No… find
  56. Rosalind W Picard. 2000. Toward computers that recognize and respond to user emotion. IBM systems journal, 39(3.4):705–719. find
  57. Robert Plutchik. 1988. The nature of emotions: Clinical implications. In Emotions and Psychopathology, pages 1–20. Springer US, Boston, MA. find
  58. Robert Plutchik. 2000. Emotions in the practice of psychotherapy: Clinical implications of affect theories. American Psychological Association. find
  59. Maria Pontiki, Dimitris Galanis, Haris Papageorgiou, Ion Androutsopoulos, Suresh Manandhar, Mohammad AL-Smadi, Mahmoud Al-Ayyoub, Yanyan Zhao, Bing Qin, Orphée De Clercq, Véronique Hoste, Marianna Apidianaki, Xavier Tannier, Natalia Loukachevitch, Evgeniy Kotelnikov, Nuria Bel, Salud María Jiménez-Z… find
  60. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. J. Mach. Learn. Res., 21(1). find
  61. Peter Salovey and John D Mayer. 1990. Emotional intelligence. Imagin. Cogn. Pers., 9(3):185–211. find
  62. Kim Schouten and Flavius Frasincar. 2016. Survey on aspect-level sentiment analysis. IEEE Transactions on Knowledge and Data Engineering, 28(3):813– 830. find
  63. Yuchen Shen and Xiaojun Wan. 2023. Opinsummeval: Revisiting automated evaluation for opinion summarization. find
  64. Tejpalsingh Siledar, Swaroop Nath, Sankara Sri Raghava Ravindra Muddu, Rupasai Rangaraju, Swaprava Nath, Pushpak Bhattacharyya, Suman Banerjee, Amey Patil, Sudhanshu Shekhar Singh, Muthusamy Chelliah, and Nikesh Garera. 2024. One prompt to rule them all: Llms for opinion summary evaluation. find
  65. Jen tse Huang, Man Ho Lam, Eric John Li, Shujie Ren, Wenxuan Wang, Wenxiang Jiao, Zhaopeng Tu, and Michael R. Lyu. 2024. Emotionally numb or empathetic? evaluating how llms feel using emotionbench. find
  66. Jiaan Wang, Yunlong Liang, Fandong Meng, Zengkui Sun, Haoxiang Shi, Zhixu Li, Jinan Xu, Jianfeng Qu, and Jie Zhou. 2023. Is ChatGPT a good NLG evaluator? a preliminary study. In Proceedings of the 4th New Frontiers in Summarization Workshop, pages 1–11, Singapore. Association for Computational Lingu… find our card
  67. Lu Wang and Wang Ling. 2016. Neural network-based abstract generation for opinions and arguments. arXiv preprint arXiv:1606.02785. find
  68. Xiting Wang, Junpeng Guo, Yi Wu, and Na Liu. 2019. Emotion as a signal of product quality: Its effect on purchase decision based on online customer reviews. Internet Research, 30(2):463–485. find
  69. James H. Steiger. 1980. Tests for comparing elements of a correlation matrix. Psychological Bulletin, 87(2):245–251. find
  70. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc Le, and Denny Zhou. 2023. Chain-of-thought prompting elicits reasoning in large language models. find our card
  71. 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 Thomas Wolf, Lysandre Debut, Victor Sanh, Julien Chaumond, Cle… find our card
  72. An Yang, Baosong Yang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Zhou, Chengpeng Li, Chengyuan Li, Dayiheng Liu, Fei Huang, Guanting Dong, Haoran Wei, Huan Lin, Jialong Tang, Jialin Wang, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Ma, Jianxin Yang, Jin Xu, Jingren Zhou, Jinze Bai, Jinzheng He, Juny… find
  73. Yusen Zhang, Nan Zhang, Yixin Liu, Alexander Fabbri, Junru Liu, Ryo Kamoi, Xiaoxin Lu, Caiming Xiong, Jieyu Zhao, Dragomir Radev, Kathleen McKeown, and Rui Zhang. 2024. Fair abstractive summarization of diverse perspectives. In Proceedings of the 2024 Conference of the North American Chapter of the… find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/attri-2026-bridging-cognition-affect-emotion-aware/