---
title: "HARPO — paper by Raj et al. (2026) — summary and ideas"
description: "They propose HARPO, an agentic framework built on an LLM backbone that treats recommendation as a decision-making… 31 references."
authors: ["Subham Raj", "Aman Vaibhav Jha", "Mayank Anand", "Sriparna Saha"]
year: 2026
venue: "Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)"
source: "https://aclanthology.org/2026.acl-long.1646/"
doi: "10.18653/v1/2026.acl-long.1646"
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/raj-2026-harpo-hierarchical-agentic-reasoning-user/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/raj-2026-harpo-hierarchical-agentic-reasoning-user/)

[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/36/) / HARPO: Hierarchical Agentic Reasoning for User-Aligned Conversational Recommendation

# HARPO: Hierarchical Agentic Reasoning for User-Aligned Conversational Recommendation

**Authors:** Subham Raj, Aman Vaibhav Jha, Mayank Anand and [Sriparna Saha](https://md.research.elamur.ai/free/en/authors/sriparna-saha/) · **Year:** 2026 · **Venue:** [Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)](https://md.research.elamur.ai/free/en/venues/acl/) ([ACL 2026](https://md.research.elamur.ai/free/en/venues/acl/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.acl-long.1646/) · [PDF](https://aclanthology.org/2026.acl-long.1646.pdf) · [DOI 10.18653/v1/2026.acl-long.1646](https://doi.org/10.18653/v1/2026.acl-long.1646) · [OpenAlex](https://openalex.org/W7166849329)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Conversational recommender systems (CRSs) usually score well on proxy measures such as Recall@K and BLEU, yet their suggestions often miss what users actually want. The authors argue that this happens because training rewards retrieval or fluent text rather than the quality of the recommendation itself. They propose HARPO, an agentic framework built on an LLM backbone that treats recommendation as a decision-making problem with explicit quality optimization. It has four parts:
- **STAR** is a tree-of-thought search guided by a value network.
- **CHARM** handles hierarchical preference learning. It splits quality into relevance, diversity, satisfaction and engagement and weights them by context.
- **BRIDGE** supports cross-domain transfer through domain-agnostic Virtual Tool Operations (VTOs).
- **MAVEN** performs multi-agent refinement.

Tested on ReDial, INSPIRED and MUSE, HARPO beats strong baselines on recommendation-focused and user-alignment metrics and stays competitive on response quality. The work matters because it pushes CRS evaluation away from proxy scores toward quality as users experience it. It also offers a tool abstraction that carries across domains.

## Key points
- Benchmarks: ReDial and INSPIRED (movies) and MUSE (multimodal fashion). Baselines include KGSF, UniCRS, GPT-4/GPT-4V, RecMind, Qwen2-VL-7B and LLaVA-Next-8B.
- On INSPIRED, HARPO's macro-averaged score is 0.454 versus 0.370 for RecMind (+22.7%). On MUSE it is 0.524 versus 0.468 for Qwen2-VL-7B (+12.0%). Reported gains are significant at p \< 0.01.
- In the ReDial ablation, the full model reaches R@10 = 29.8. Removing VTOs drops it to 23.4, removing CHARM to 24.6, and SFT-only training gives 21.6.
- The learned reward dimensions correlate with matching human ratings at r = 0.64–0.73. Context shifts the weights; for example, the diversity weight rises to 0.31 when a user asks for novelty.
- Efficiency: a beam width of 3 delivers about 99% of the performance of width 5 with 40% lower latency. LoRA rank 16 is sufficient.
- Stated limitations: the four-stage training is costly, the VTO taxonomy is designed by hand, and human evaluation is limited.

*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

- [Indian Institute of Technology Patna](https://md.research.elamur.ai/free/en/universities/indian-institute-of-technology-patna/) — University, Patna, India · 21 works · 374 citations
- [Indian Institute of Information Technology Allahabad](https://md.research.elamur.ai/free/en/universities/indian-institute-of-information-technology-allahabad/) — University, Prayagraj, India · 2 works · 11 citations

📚 Shelves (4): [🖥️ Computer Science](https://md.research.elamur.ai/free/en/shelves/computer-science/) · 18,614 works · [🗄️ Information Systems](https://md.research.elamur.ai/free/en/shelves/computer-science/information-systems/) · 1,730 works · [🎯 Decision Sciences](https://md.research.elamur.ai/free/en/shelves/decision-sciences/) · 249 works · [🎯 Management Science and Operations Research](https://md.research.elamur.ai/free/en/shelves/decision-sciences/management-science-and-operations-research/) · 134 works

🧭 Topics (2): [🧮 Reasoning](https://md.research.elamur.ai/free/en/topics/reasoning/) · 1,685 works · [🗨️ Dialogue systems](https://md.research.elamur.ai/free/en/topics/dialogue-systems/) · 654 works

## Cite

Cite as: Subham Raj, Aman Vaibhav Jha, Mayank Anand and Sriparna Saha. 2026. HARPO: Hierarchical Agentic Reasoning for User-Aligned Conversational Recommendation. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

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

```bibtex
@inproceedings{raj2026harpo,
  title = {{HARPO: Hierarchical Agentic Reasoning for User-Aligned Conversational Recommendation}},
  author = {Subham Raj, Aman Vaibhav Jha, Mayank Anand and Sriparna Saha},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.1646},
  url = {https://doi.org/10.18653/v1/2026.acl-long.1646}
}
```

- [← previous Harnessing the Latent Space: From Steering Vectors to Model Calibrators for Control and Trust](https://md.research.elamur.ai/free/en/papers/subramani-2026-harnessing-latent-space-steering-vectors/)
- [next → Hate Speech Detection in Turkish and Arabic: A Comprehensive Study](https://md.research.elamur.ai/free/en/papers/dehghan-2026-hate-speech-detection-turkish-arabic/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/36/)

## References

31 references, 9 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Qibin Chen, Junyang Lin, Yichang Zhang, Ming Ding, Yukuo Cen, Hongxia Yang, and Jie Tang. 2019. Towards knowledge-based recommender dialog system. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language… [find](https://scholar.google.com/scholar?q=Qibin+Chen%2C+Junyang+Lin%2C+Yichang+Zhang%2C+Ming+Ding%2C+Yukuo+Cen%2C+Hongxia+Yang%2C+and+Jie+Tang%2E+2019%2E+Towards+knowledge%2Dbased+recommender+dialog+system%2E+In+Proceedings+of+the+2019+Conference+on+Empirical+Me)
2. Konstantina Christakopoulou, Filip Radlinski, and Katja Hofmann. 2016. Towards conversational recommender systems. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 815–824. [find](https://scholar.google.com/scholar?q=Konstantina+Christakopoulou%2C+Filip+Radlinski%2C+and+Katja+Hofmann%2E+2016%2E+Towards+conversational+recommender+systems%2E+In+Proceedings+of+the+22nd+ACM+SIGKDD+International+Conference+on+Knowledge+Discovery)
3. Huy Dao, Yang Deng, Dung D Le, and Lizi Liao. 2024. Broadening the view: Demonstration-augmented prompt learning for conversational recommendation. In Proceedings of the 47th International ACM SI-GIR Conference on Research and Development in Information Retrieval, pages 785–795. [find](https://scholar.google.com/scholar?q=Huy+Dao%2C+Yang+Deng%2C+Dung+D+Le%2C+and+Lizi+Liao%2E+2024%2E+Broadening+the+view%3A+Demonstration%2Daugmented+prompt+learning+for+conversational+recommendation%2E+In+Proceedings+of+the+47th+International+ACM+SI%2DGIR+)
4. Tri Dao. 2023. [Flashattention-2: Faster attention with better parallelism and work partitioning](https://md.research.elamur.ai/free/en/papers/dao-2023-flashattention-2-faster-attention-better/). arXiv preprint arXiv:2307.08691.
5. Yaroslav Ganin and Victor Lempitsky. 2015. Unsupervised domain adaptation by backpropagation. In International conference on machine learning, pages 1180–1189. PMLR. [find](https://scholar.google.com/scholar?q=Yaroslav+Ganin+and+Victor+Lempitsky%2E+2015%2E+Unsupervised+domain+adaptation+by+backpropagation%2E+In+International+conference+on+machine+learning%2C+pages+1180%E2%80%931189%2E+PMLR%2E)
6. Chongming Gao, Wenqiang Lei, Xiangnan He, Maarten de Rijke, and Tat-Seng Chua. 2021. Advances and challenges in conversational recommender systems: A survey. AI Open, 2:100–126. [find](https://scholar.google.com/scholar?q=Chongming+Gao%2C+Wenqiang+Lei%2C+Xiangnan+He%2C+Maarten+de+Rijke%2C+and+Tat%2DSeng+Chua%2E+2021%2E+Advances+and+challenges+in+conversational+recommender+systems%3A+A+survey%2E+AI+Open%2C+2%3A100%E2%80%93126%2E)
7. Shirley Anugrah Hayati, Dongyeop Kang, Qingxi-aoyang Zhu, Weiyan Shi, and Zhou Yu. 2020. Inspired: Toward sociable recommendation dialog systems. In Proceedings of the 2020 conference on empirical methods in natural language processing (EMNLP), pages 8142–8152. [find](https://scholar.google.com/scholar?q=Shirley+Anugrah+Hayati%2C+Dongyeop+Kang%2C+Qingxi%2Daoyang+Zhu%2C+Weiyan+Shi%2C+and+Zhou+Yu%2E+2020%2E+Inspired%3A+Toward+sociable+recommendation+dialog+systems%2E+In+Proceedings+of+the+2020+conference+on+empirical+met)
8. Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Liang Wang, Weizhu Chen, et al. 2022. [Lora: Low-rank adaptation of large language models](https://md.research.elamur.ai/free/en/papers/hu-2021-lora-low-rank-adaptation-large/). Iclr, 1(2):3.
9. Xu Huang, Jianxun Lian, Yuxuan Lei, Jing Yao, Defu Lian, and Xing Xie. 2025. Recommender ai agent: Integrating large language models for interactive recommendations. ACM Transactions on Information Systems, 43(4):1–33. [find](https://scholar.google.com/scholar?q=Xu+Huang%2C+Jianxun+Lian%2C+Yuxuan+Lei%2C+Jing+Yao%2C+Defu+Lian%2C+and+Xing+Xie%2E+2025%2E+Recommender+ai+agent%3A+Integrating+large+language+models+for+interactive+recommendations%2E+ACM+Transactions+on+Information+Sy)
10. Dietmar Jannach, Ahtsham Manzoor, Wanling Cai, and Li Chen. 2021. A survey on conversational recommender systems. ACM Computing Surveys, 54(5):1– 36. [find](https://scholar.google.com/scholar?q=Dietmar+Jannach%2C+Ahtsham+Manzoor%2C+Wanling+Cai%2C+and+Li+Chen%2E+2021%2E+A+survey+on+conversational+recommender+systems%2E+ACM+Computing+Surveys%2C+54%285%29%3A1%E2%80%93+36%2E)
11. Walid Krichene and Steffen Rendle. 2020. On sampled metrics for item recommendation. In Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, pages 1748– 1757. [find](https://scholar.google.com/scholar?q=Walid+Krichene+and+Steffen+Rendle%2E+2020%2E+On+sampled+metrics+for+item+recommendation%2E+In+Proceedings+of+the+26th+ACM+SIGKDD+international+conference+on+knowledge+discovery+%26+data+mining%2C+pages+1748%E2%80%93+17)
12. Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. 2023. [Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models](https://md.research.elamur.ai/free/en/papers/li-2023-blip-2-bootstrapping-language-image/). In International conference on machine learning, pages 19730–19742. PMLR.
13. Raymond Li, Samira Ebrahimi Kahou, Hannes Schulz, Vincent Michalski, Laurent Charlin, and Chris Pal. 2018. Towards deep conversational recommendations. Advances in neural information processing systems, 31. [find](https://scholar.google.com/scholar?q=Raymond+Li%2C+Samira+Ebrahimi+Kahou%2C+Hannes+Schulz%2C+Vincent+Michalski%2C+Laurent+Charlin%2C+and+Chris+Pal%2E+2018%2E+Towards+deep+conversational+recommendations%2E+Advances+in+neural+information+processing+system)
14. Prabir Mondal, Pulkit Kapoor, Siddharth Singh, Sriparna Saha, Jyoti Prakash Singh, and Amit Kumar Singh. 2023. Genre effect toward developing a multimodal movie recommendation system in indian setting. IEEE Transactions on Consumer Electronics, 70(1):2517–2526. [find](https://scholar.google.com/scholar?q=Prabir+Mondal%2C+Pulkit+Kapoor%2C+Siddharth+Singh%2C+Sriparna+Saha%2C+Jyoti+Prakash+Singh%2C+and+Amit+Kumar+Singh%2E+2023%2E+Genre+effect+toward+developing+a+multimodal+movie+recommendation+system+in+indian+setting)
15. Long Ouyang, Jeffrey Wu, Xu Jiang, Diogo Almeida, Carroll Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, et al. 2022. [Training language models to follow instructions with human feedback](https://md.research.elamur.ai/free/en/papers/ouyang-2022-training-language-models-follow-instructions/). Advances in neural information processing systems, 35:27730–27744.
16. Yujia Qin, Shihao Liang, Yining Ye, Kunlun Zhu, Lan Yan, Yaxi Lu, Yankai Lin, Xin Cong, Xiangru Tang, Bill Qian, et al. 2023. [Toolllm: Facilitating large language models to master 16000+ real-world apis](https://md.research.elamur.ai/free/en/papers/qin-2023-toolllm-facilitating-large-language-models/). arXiv preprint arXiv:2307.16789.
17. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D Manning, Stefano Ermon, and Chelsea Finn. 2023. [Direct preference optimization: Your language model is secretly a reward model](https://md.research.elamur.ai/free/en/papers/rafailov-2023-direct-preference-optimization-language-model/). Advances in neural information processing systems, 36:53728–53741.
18. Subham Raj, Prabir Mondal, Daipayan Chakder, Sriparna Saha, and Naoyuki Onoe. 2023. A multi-modal multi-task based approach for movie recommendation. In 2023 International Joint Conference on Neural Networks (IJCNN), pages 1–8. IEEE. [find](https://scholar.google.com/scholar?q=Subham+Raj%2C+Prabir+Mondal%2C+Daipayan+Chakder%2C+Sriparna+Saha%2C+and+Naoyuki+Onoe%2E+2023%2E+A+multi%2Dmodal+multi%2Dtask+based+approach+for+movie+recommendation%2E+In+2023+International+Joint+Conference+on+Neural+N)
19. Subham Raj, Sriparna Saha, Brijraj Singh, and Niranjan Pedanekar. 2025. Multimodal movie recommendation with multitasking architecture and learning user– movie representation: An empirical study. IEEE Transactions on Computational Social Systems. [find](https://scholar.google.com/scholar?q=Subham+Raj%2C+Sriparna+Saha%2C+Brijraj+Singh%2C+and+Niranjan+Pedanekar%2E+2025%2E+Multimodal+movie+recommendation+with+multitasking+architecture+and+learning+user%E2%80%93+movie+representation%3A+An+empirical+study%2E+IEEE)
20. Diederik M Roijers, Peter Vamplew, Shimon Whiteson, and Richard Dazeley. 2013. A survey of multi-objective sequential decision-making. Journal of Artificial Intelligence Research, 48:67–113. [find](https://scholar.google.com/scholar?q=Diederik+M+Roijers%2C+Peter+Vamplew%2C+Shimon+Whiteson%2C+and+Richard+Dazeley%2E+2013%2E+A+survey+of+multi%2Dobjective+sequential+decision%2Dmaking%2E+Journal+of+Artificial+Intelligence+Research%2C+48%3A67%E2%80%93113%2E)
21. Timo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu, Maria Lomeli, Eric Hambro, Luke Zettlemoyer, Nicola Cancedda, and Thomas Scialom. 2023. [Toolformer: Language models can teach themselves to use tools](https://md.research.elamur.ai/free/en/papers/schick-2023-toolformer-language-models-can-teach/). Advances in neural information processing systems, 36:68539–68551.
22. Yueming Sun and Yi Zhang. 2018. Conversational recommender system. In The 41st international acm sigir conference on research & development in information retrieval, pages 235–244. [find](https://scholar.google.com/scholar?q=Yueming+Sun+and+Yi+Zhang%2E+2018%2E+Conversational+recommender+system%2E+In+The+41st+international+acm+sigir+conference+on+research+%26+development+in+information+retrieval%2C+pages+235%E2%80%93244%2E)
23. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. 2023. [Llama 2: Open foundation and fine-tuned chat models](https://md.research.elamur.ai/free/en/papers/touvron-2023-llama-2-open-foundation-fine/). arXiv preprint arXiv:2307.09288.
24. Ting-Chun Wang, Shang-Yu Su, and Yun-Nung Chen. 2022a. Barcor: Towards a unified framework for conversational recommendation systems. arXiv preprint arXiv:2203.14257. — [arXiv:2203.14257](https://arxiv.org/abs/2203.14257) [find](https://scholar.google.com/scholar?q=Ting%2DChun+Wang%2C+Shang%2DYu+Su%2C+and+Yun%2DNung+Chen%2E+2022a%2E+Barcor%3A+Towards+a+unified+framework+for+conversational+recommendation+systems%2E+arXiv+preprint+arXiv%3A2203%2E14257%2E+%E2%80%94+%5BarXiv%3A2203%2E14257%5D%28https%3A%2F%2Farxi)
25. Xiaolei Wang, Kun Zhou, Ji-Rong Wen, and Wayne Xin Zhao. 2022b. Towards unified conversational recommender systems via knowledge-enhanced prompt learning. In Proceedings of the 28th ACM SIGKDD conference on knowledge discovery and data mining, pages 1929–1937. [find](https://scholar.google.com/scholar?q=Xiaolei+Wang%2C+Kun+Zhou%2C+Ji%2DRong+Wen%2C+and+Wayne+Xin+Zhao%2E+2022b%2E+Towards+unified+conversational+recommender+systems+via+knowledge%2Denhanced+prompt+learning%2E+In+Proceedings+of+the+28th+ACM+SIGKDD+confere)
26. Yancheng Wang, Ziyan Jiang, Zheng Chen, Fan Yang, Yingxue Zhou, Eunah Cho, Xing Fan, Yanbin Lu, Xiaojiang Huang, and Yingzhen Yang. 2024. Recmind: Large language model powered agent for recommendation. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 4351–4364. [find](https://scholar.google.com/scholar?q=Yancheng+Wang%2C+Ziyan+Jiang%2C+Zheng+Chen%2C+Fan+Yang%2C+Yingxue+Zhou%2C+Eunah+Cho%2C+Xing+Fan%2C+Yanbin+Lu%2C+Xiaojiang+Huang%2C+and+Yingzhen+Yang%2E+2024%2E+Recmind%3A+Large+language+model+powered+agent+for+recommendation)
27. Zihan Wang, Xiaocui Yang, Yongkang Liu, Shi Feng, Daling Wang, and Yifei Zhang. 2025. Muse: A multimodal conversational recommendation dataset with scenario-grounded user profiles. In Findings of the Association for Computational Linguistics: ACL 2025, pages 1027–1053. [find](https://scholar.google.com/scholar?q=Zihan+Wang%2C+Xiaocui+Yang%2C+Yongkang+Liu%2C+Shi+Feng%2C+Daling+Wang%2C+and+Yifei+Zhang%2E+2025%2E+Muse%3A+A+multimodal+conversational+recommendation+dataset+with+scenario%2Dgrounded+user+profiles%2E+In+Findings+of+the+)
28. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Brian Ichter, Fei Xia, Ed Chi, Quoc 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 Advances in Neural Information Processing Systems, volume 35, pages 24824–24837.
29. Yibiao Wei, Jie Zou, Weikang Guo, Guoqing Wang, Xing Xu, and Yang Yang. 2025. Mscrs: Multi-modal semantic graph prompt learning framework for conversational recommender systems. In Proceedings of the 48th International ACM SIGIR Conference on Research and Development in Information Retrieval, pages… [find](https://scholar.google.com/scholar?q=Yibiao+Wei%2C+Jie+Zou%2C+Weikang+Guo%2C+Guoqing+Wang%2C+Xing+Xu%2C+and+Yang+Yang%2E+2025%2E+Mscrs%3A+Multi%2Dmodal+semantic+graph+prompt+learning+framework+for+conversational+recommender+systems%2E+In+Proceedings+of+the+)
30. Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Tom Griffiths, Yuan Cao, and Karthik Narasimhan. 2023. Tree of thoughts: Deliberate problem solving with large language models. Advances in neural information processing systems, 36:11809–11822. [find](https://scholar.google.com/scholar?q=Shunyu+Yao%2C+Dian+Yu%2C+Jeffrey+Zhao%2C+Izhak+Shafran%2C+Tom+Griffiths%2C+Yuan+Cao%2C+and+Karthik+Narasimhan%2E+2023%2E+Tree+of+thoughts%3A+Deliberate+problem+solving+with+large+language+models%2E+Advances+in+neural+inf)
31. Kun Zhou, Wayne Xin Zhao, Shuqing Bian, Yuanhang Zhou, Ji-Rong Wen, and Jingsong Yu. 2020. Improving conversational recommender systems via knowledge graph based semantic fusion. In Proceedings of the 26th ACM SIGKDD international conference on knowledge discovery & data mining, pages 1006– 1014. [find](https://scholar.google.com/scholar?q=Kun+Zhou%2C+Wayne+Xin+Zhao%2C+Shuqing+Bian%2C+Yuanhang+Zhou%2C+Ji%2DRong+Wen%2C+and+Jingsong+Yu%2E+2020%2E+Improving+conversational+recommender+systems+via+knowledge+graph+based+semantic+fusion%2E+In+Proceedings+of+the)

*Updated 1 November 2026*