See you in Nanjing on 19–21 October!
Registration is still open (regular fee, on-site or online participation). Registration details - Registration form
The programme is available (PDF).
The proceedings are published by Springer (LNAI 17186).
Welcome reception on 19 October and gala dinner on 20 October.
The BELIEF 2026 proceedings, Belief Functions: Theory and Applications (LNAI 17186), are now available from Springer.
The programme is now available, as well as accommodation and travel information.
Elsevier will sponsor an IJAR Best Paper Award, which will be presented to an early-career researcher.
The Theory of belief functions, also referred to as Evidence theory or Dempster-Shafer theory is a well-established mathematical framework for reasoning and decision-making with uncertainty, with applications to machine learning, statistical inference, information fusion, knowledge representation, risk analysis, etc. It is an extension of probability theory and it has well-understood connections with related frameworks such as random sets and imprecise probabilities. Recently, generalised evidence theory, based on random fuzzy sets, has been proposed as a unifying framework encompassing both Dempster-Shafer and possibility theories.
The biennial BELIEF conferences (sponsored by the Belief Functions and Applications Society) are dedicated to the confrontation of ideas, the reporting of recent achievements and the presentation of the wide range of applications of this theory. Previous editions of this conference series were held in Brest, France (2010); Compiègne, France (2012); Oxford, UK (2014); Prague, Czech Republic (2016); Compiègne, France (2018); Shanghai, China (2021); Paris, France (2022); and Belfast, UK (2024).
To support cross-fertilisation among researchers working in different subfields of AI and related disciplines, tutorials and special sessions will be proposed, with emphasis on the links between machine learning, statistical inference and uncertain reasoning, including topics such as quantification of prediction uncertainty, fusion rules for ensemble learning, belief propagation in deep neural networks, links with explainable and symbolic AI, etc. Submissions of papers combining several of these topics, or more generally at the crossroads of belief functions and other AI methods or uncertainty theories, along with relevant applications, are welcome.
The Ninth International Conference on Belief Functions (BELIEF 2026) will be held in Nanjing, China, on 19–21 October 2026.
The BELIEF 2026 proceedings are published by Springer in the Lecture Notes in Artificial Intelligence (LNCS/LNAI) series, volume 17186: Belief Functions: Theory and Applications, edited by Thierry Denœux, Anne-Laure Jousselme and Zhi-gang Su.
Authors of selected papers from the BELIEF 2026 conference will be invited to submit extended versions of their papers for possible inclusion in a special issue of the International Journal of Approximate Reasoning.
Authors should submit their papers through Microsoft CMT following Springer's submission guidelines.
Papers should be no longer than 8 pages, references included, and should present original contributions with significant results.
The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.
Original contributions are solicited on theoretical aspects, but not limited to
as well as on applications to various areas including, but not limited to
Each paper must have at least one author registered with an academic/industrial registration (i.e., not a student registration) with payment received by the author registration deadline to avoid being withdrawn from the conference. A single, academic/industrial registration may cover up to two (2) papers. When registering, please enter the ID number(s) of the paper(s) that you will be covering with your academic/industrial registration.
If you are a student who will be presenting a paper, and you are covering that paper with your registration, then an academic/industrial registration is required for you.
For each accepted paper, BELIEF 2026 requires a presentation by one of the authors, either in person or online if registered for online participation.
Any paper accepted in the technical programme, but not presented on-site or online, will be removed from the official proceedings archived by Springer.
Final versions of accepted papers must strictly adhere to the Springer LNCS templates available here or on Overleaf at Springer LNCS Overleaf. Papers must not exceed 8 pages, with up to 2 additional pages for references. Springer encourages authors to include their ORCID iDs in their papers. Authors should also comply with the Springer Code of Conduct. Please submit the following materials via CMT by 31 July 2026:
The BELIEF 2026 Conference will take place at the Sipailou Campus of Southeast University, Nanjing, China.
Address: 2 Sipailou, Southeast University, Xuanwu District, Nanjing, Jiangsu, China, 210018
Map search: "SEU Sipailou Campus - Zhongshan College". Please note that in China, the Baidu Maps or Gaode (Amap) apps work better than Google Maps.
Nanjing, capital of Jiangsu Province in eastern China, is a historic city with over 2,500 years of history, and was once the capital of several dynasties, including the Ming Dynasty. Renowned for its cultural heritage, it is home to iconic sites such as the old city walls, the Temple of Confucius and the Sun Yat-sen Mausoleum. Today, Nanjing is a modern metropolis with a thriving economy, leading universities and a blend of traditional and contemporary architecture. Its strategic location along the Yangtze River has long made it a key centre for trade and culture.
Please note that October is the peak tourist season in Nanjing. Please book your hotel at least three weeks in advance.
For participants not familiar with Chinese, we recommend using Booking.com (in English) to book one of the following hotels:
Price: about 584 RMB (75€) per night, breakfast included
Distance to the venue: about 7 minutes on foot
Address: No. 389, Zhujiang Road, Xuanwu District, 210018 Nanjing, China
Price: about 349 RMB (42€) per night, breakfast included
Distance to the venue: about 12 minutes on foot, or about 3 minutes by shared bicycle
Address: No. 39, North Taiping Road, Xuanwu District, 210000 Nanjing, China
Price: about 305 RMB (39€) per night, breakfast included
Distance to the venue: about 20 minutes on foot, or about 5 minutes by shared bicycle
Address: No. 29-1, Longpan Middle Road, Xuanwu District, 210000 Nanjing, China
10月是南京旅游旺季，请尽可能提前预定酒店。
价格：550人民币/晚
地址：南京玄武区珠江路389号
距离：步行10分钟
价格：300人民币/晚
地址：南京玄武区大石桥街29号
距离：步行10分钟
价格：150人民币/晚
地址：南京玄武区四牌楼61号2楼
距离：步行4分钟
1. Subway (recommended)
2. Bus
Subway (recommended)
| Format | Single-track conference with on-site and online participation |
|---|---|
| Time zone | China Standard Time (UTC+8) |
| Contributed talks | 15-minute presentation followed by 5 minutes of discussion |
| Conference start | Monday 19 October at 08:30 (registration); opening at 09:00 |
| Conference close | Wednesday 21 October at approximately 16:00 |
| Date | Morning | Afternoon | Evening |
|---|---|---|---|
| Mon 19 Oct | Opening, Keynote 1, Theory | Deep Learning | Welcome Reception |
| Tue 20 Oct | Keynote 2, Classification | Lifetime Prediction, Decision Analysis, BFAS Assembly | Gala Dinner and Best Paper Award |
| Wed 21 Oct | Large Language Models, Information Fusion | Clustering and Closing | – |
Detailed room assignments and online access information will be announced later.
| 08:30–09:00 | Registration |
| 09:00–09:20 | Opening Ceremony |
| 09:20–10:20 | Keynote 1: Optimal transport under Dempster-Shafer theory (Chair: Thierry Denœux) Barbara Vantaggi, Sapienza University of Rome, Italy |
| 10:20–10:40 | Coffee Break |
| 10:40–12:40 | Session 1: Theory (Chair: Frédéric Pichon) |
| 10:40–11:00 | Walsh-Hadamard Transform of Basic Probability Assignment Ruolan Cheng, Yong Deng |
| 11:00–11:20 | Dempster, Geometric and Bayesian Conditionings w.r.t. Random Sets Andrey Bronevich, Igor Rozenberg |
| 11:20–11:40 | Gamma-based Distance Measure For Layer-2 Belief Structure Luyuan Chen, Xinghua Zhou, Qianli Zhou |
| 11:40–12:00 | The variety of evidence thesis holds in evidence theory Frédéric Pichon, Juergen Landes, Sébastien Destercke |
| 12:00–12:20 | Order-N Entropies as Conflict Measures in Generalised Evidence Theory Thierry Denœux, Anne-Laure Jousselme, Frédéric Pichon |
| 12:20–12:40 | Combining Reliability Judgments Liping Liu |
| 12:40–14:00 | Lunch |
| 14:00–16:00 | Session 2: Deep Learning (Chair: Xinyang Deng) |
| 14:00–14:20 | Neural Networks with Differentiable Belief-Function Layers Antonin Sulc |
| 14:20–14:40 | Deep Belief Forest for Pattern Classification Deqiang Han, Wei Li, Jean Dezert |
| 14:40–15:00 | Deep Learning Uncertainty Qualification with Belief Structure Qianli Zhou, Mingxi Yin, Xinyang Deng, Wen Jiang |
| 15:00–15:20 | Ev-YOLO: Uncertainty-Aware Object Detection via a Unified Evidential Formulation Simon Barbarit-Gaboriau, Hind Laghmara, Rémi Boutteau, Samia Ainouz |
| 15:20–15:40 | Evidential Feature Maps for Semantic Scene Segmentation Abdelhak Benamirouche, Hind Laghmara, Rémi Boutteau, Jean-Philippe Lauffenburger |
| 15:40–16:00 | Uncertainty-Aware Deep Learning for Breast Cancer Detection in Mammograms with Breast Implants Junjie Huang, Yucheng Xing, Ling Huang, Hao Du |
| 16:00–16:20 | Coffee Break |
| 17:30–19:00 | Welcome Reception Venue to be announced |
| 09:00–10:00 | Keynote 2: Pattern classification with belief functions and deep neural networks (Chair: Zhi-gang Su) Zhunga Liu, Northwestern Polytechnical University, China |
| 10:00–10:20 | Coffee Break |
| 10:20–12:20 | Session 3: Classification (Chair: Didier Coquin) |
| 10:20–10:40 | QEKNN: Quantum Evidential K-Nearest Neighbors Classifier for Uncertain Pattern Classification Ruihan Li, Junjie Huang, Fuyuan Xiao, Xiaoling Li |
| 10:40–11:00 | Mixture of Evidential Experts: Differentiable Dempster-Shafer Fusion for Robust Classification Toan Nguyen-Mau, Hideomi Gokon, Van-Nam Huynh |
| 11:00–11:20 | Evidential Counterfactual Explanations for Instance-Based Classification under the Belief Function Framework Abderrahmane Chelouche, Safa Ben Ayed, Zied Elouedi, Eric Lefèvre |
| 11:20–11:40 | MassMatch: Semi-Supervised Learning through Dempster Shafer Theory Han Lam, Toan Nguyen-Mau, Duc-Vinh Vo, Van-Nam Huynh |
| 11:40–12:00 | Continual Novel Target Class Identification in Open-World Frame of Discernment Xuan Zhao, Zhikun Feng, Ping Kuang |
| 12:00–12:20 | Conditional Dempster-Shafer Modeling for Avalanche Risk Assessment Didier Coquin, Frédéric Pourraz, Gilles Mauris |
| 12:20–13:40 | Lunch |
| 13:40–14:40 | Session 4: Lifetime Prediction (Chair: Thierry Denœux) |
| 13:40–14:00 | Remaining Useful Life Prediction and Its Uncertainty Quantification Using Gaussian Random Fuzzy Numbers Mingyu Liao, Ling Huang, Tangfan Xiahou |
| 14:00–14:20 | Uncertainty Quantification for Lithium-ion Battery RUL Prediction Using Deep Evidential Regression Haoyu Fang, Yuhan Wang, Ziheng Ni, Anxu Chen, Kai Wang, Junqin Yan, Xianyong Peng |
| 14:20–14:40 | CR-ENNreg: An Evidential Survival Model for Competing Risks Analysis Chunlin Mi, Su Ruan, Ling Huang, Pierre Vera, Thierry Denœux |
| 14:40–15:00 | Coffee Break |
| 15:00–16:20 | Session 5: Decision Analysis and Hypothesis Testing (Chair: Chunlai Zhou) |
| 15:00–15:20 | Evidential Reasoning-Based Mode-Combination Selection for Nonstationary Signal Denoising Xinyi Zhang, Shiyi Su |
| 15:20–15:40 | A Semantic Association Model-Based Decision-Making Method Considering Correlations among Assessment Grades Xinyang Deng, Xiang Li, Qianli Zhou, Wen Jiang |
| 15:40–16:00 | Decision-Aware Approximation of Belief Functions for Evidential Combinatorial Optimization Sohaib Afifi |
| 16:00–16:20 | Two Evidential Frameworks for Differential Privacy Ning Liu, Hui Zou, Chunlai Zhou |
| 16:30–17:00 | BFAS General Assembly Hybrid participation for BFAS members; online access details to be announced |
| 18:30–20:30 | Gala Dinner Venue to be announced; Best Paper Award presentation during dinner |
| 09:00–10:20 | Session 6: Large Language Models (Chair: Zheng Tong) |
| 09:00–09:20 | Typographic Attack Detection in Large Vision-Language Models via Attention-Level Evidential Conflict Xiaofei Liu, Tao Huang, Rui Wang, Liping Jing |
| 09:20–09:40 | An Evidential Fusion Method for LLM-Based Multi-Agent Consensus-Reaching Yuxiang Bao, Zheng Tong |
| 09:40–10:00 | Source-Aware Belief Cues and Evidence Weighting in Clinical large language model Shuai Luo, Ke Zou, Ling Huang, Meng Liu, Nan Zhou, Jian Zhong, Lijun Jiang, Diwen Zhang, Lanxiang Wang, Yi Zhang, Hu Chen |
| 10:00–10:20 | Event-Triggered Belief Consensus over Large Language Model-Enabled Multi-Agent System with Dynamic Topologies Zheng Tong, Yuxiang Bao |
| 10:20–10:40 | Coffee Break |
| 10:40–12:40 | Session 7: Information Fusion (Chair: Fuyuan Xiao) |
| 10:40–11:00 | A Belief Function-Based Multi-Source Fusion Method for Uncertainty-Aware Short-Term Traffic State Prediction Wei Bai, Boya Yang, Weijie Wang, Linheng Li |
| 11:00–11:20 | Belief-Function-Based Evidence Fusion for Uncertainty-Aware Mechanism Inference of Erigeron breviscapus in Ethnic Medicine Siqi Li, Ling Huang |
| 11:20–11:40 | Monte Carlo Combination of Random Fuzzy Sets Thierry Denœux |
| 11:40–12:00 | Apple Grading Based on the Combination of Dependent Gaussian Random Fuzzy Numbers Yuhang Yao, Mingxiao Xin, Liyao Ma, Shuhui Bi |
| 12:00–12:20 | Confidence-Aware Evidential Fusion for Multi-Modal Drone Detection Xiangqiong Xiao, Chunzhao Ma, Lianmeng Jiao |
| 12:20–12:40 | Evidential Full-State-Graph-Assisted Fusion for Trustworthy Gas Turbine Power Prediction Zhiqian He, Haoyu Fang, Jinwei Chen |
| 12:40–14:00 | Lunch |
| 14:00–15:20 | Session 8: Clustering (Chair: Chaoyu Gong) |
| 14:00–14:20 | KCECM: A Kernel-Based Constrained Evidential C-Means Clustering Method Feng Li, Ling Hou, Jusheng Mi, Yanmeng Wang |
| 14:20–14:40 | Reliability-Stratified Evidential Clustering with Structural Anchors Haiyue Fu, Qian Wang, Zhi-gang Su |
| 14:40–15:00 | Automatic Belief Peak Selection for Incomplete Data Clustering Chaoyu Gong, Tianyi Huang, Qian Wang |
| 15:00–15:20 | Evidential Cluster-Based Active Learning Nawres Alibi, Zied Elouedi, Eric Lefèvre |
| 15:20–15:40 | Coffee Break |
| 15:40–16:00 | Closing Ceremony |
Northwestern Polytechnical University, China
Pattern classification with belief functions and deep neural networks
| Type | Early bird (closed on 31 August 2026) | Registration (on-site or online) |
|---|---|---|
| Academic/industrial (Regular) | 2600 RMB (330€) | |
| Student | 1800 RMB (230€) |
The early bird period has ended: the regular registration fee now applies to all registrations, on-site or online.
| Registration fees cover: |
| Full on-site or online participation to the BELIEF 2026 conference, including keynote presentations |
| BELIEF 2026 conference proceedings |
| Gala dinner |
| Welcome reception |
| Coffee breaks and lunches |
| BFAS membership for 4 years |
Participants have to pay the suitable registration fee. Before proceeding, please read the payment guidelines carefully:
The Belief Functions and Applications Society (BFAS) offered a number of grants, covering the registration fees, to allow students with limited funding opportunities to present their work at the conference (on-site or online). The application period closed on 30 June 2026 and the grants have been awarded; applicants were notified in July.
Participants are requested to familiarise themselves with the applicable visa requirements well in advance of the conference. Present worldwide security regulations generally result in more stringent visa requirements and associated longer visa processing times. The BELIEF 2026 organisers cannot intervene with embassies or consulates on behalf of any participant.
In order to obtain a letter of invitation for visa purposes, contact the Organising Committee directly.