Supported by

Ninth International Workshop on Symbolic-Neural Learning (SNL2025)

October 29-30, 2025
Nakanoshima Center 10F, The University of Osaka (Osaka, Japan)

Program

October 29 (Wednesday)
13:00- 13:10	Opening
13:10- 14:00	Keynote Talk I	Yun-Nung (Vivian) Chen	Strategizing Conversations: Reasoning for Personalized AI Agents
14:00- 14:50	Keynote talk II	Katsushi Ikeuchi	Learning-from-Observation2.0
14:50- 15:50	Poster Session
15:50- 16:20	Coffee Break
16:20- 16:45	Invited Talk Ⅰ	Gou Koutaki	Ensemble System Using Semi-Automatic Instrument-Playing Robots
16:45- 17:10	Invited Talk Ⅱ	Sho Sonoda	Conjecturing-Proving Loop: Discovering New Theorems via LLMs with In-Context Proof Learning in Lean
17:10- 17:35	Invited Talk Ⅲ	Yusuke Matsui	Where Learned Data Structures Meet Computer Vision
17:50- 19:50	Banquet		@Nakanoshima Center
October 30 (Thursday)
10:00- 10:50	Keynote Talk Ⅲ	David Chiang (online)	What Transformers Can and Can't Do: A Logical Approach
10:50- 11:00	Break
11:00- 11:25	Invited Talk Ⅳ	Mayumi Bono	Improvisational Signing: How Deaf People Orient to and Engage with Symbolic Resources
11:25- 11:50	Invited Talk V	Takashi Matsubara	How First-order Logic Helps Diffusion-based Image Generation
11:50- 13:20	Lunch Break		- Lunch map (little bit out-dated) - Around Fukushima station (Google Maps) - Around Nakanoshima (Google Maps)
13:20- 13:45	Invited Talk Ⅵ	Shuhei Kurita	Real-world foundation models: from Text toward Egocentric-vision, 3D and Robotics
13:45- 14:10	Invited Talk Ⅶ	Shinnosuke Takamichi	How Do Audio Foundation Models Understand Sound?
14:10- 14:35	Invited Talk Ⅷ	Tatsuya Yokota	Tensor Network Decompositions and Their Applications in Machine Learning
14:35- 15:00	Coffee Break
15:00- 15:50	Keynote Talk Ⅳ	Kyle Richardson	Understanding the Logic of Generative AI through Logic and Programming
15:50- 16:00	Break
16:00- 16:25	Invited Talk Ⅸ	Naoto Yokoya	Open and Equitable AI for Earth Observation
16:25- 17:15	Keynote Talk Ⅴ	Matt Walter	From Representations to Policies: Neural-Symbolic Robot Learning from Demonstrations
17:15- 17:25	Closing

Poster Session

(P01) UCB-Guided Diffusion with Self-Prediction Training for Text Generation

Masaki Asada (National Institute of Advanced Industrial Science and Technology), Makoto Miwa (Toyota Technological Institute/National Institute of Advanced Industrial Science and Technology)

(P02) Regularizing Supervised Discriminative Learning with Diffusion Models

Takuya Asakura (Institute of Science Tokyo), Nakamasa Inoue (Institute of Science Tokyo), Koichi Shinoda (Institute of Science Tokyo)

(P03) FreeEyeglass: Training-free and Mask-free Eyeglass Transfer for Facial Videos

Weng Ian Chan (The University of Osaka), Yuantian Huang (CyberAgent AI Lab), Xingchao Yang (CyberAgent AI Lab), Fumio Okura (The University of Osaka), Takafumi Taketomi (CyberAgent AI Lab)

(P04) MedAgents: A Coordinated Multi-Agent Framework for Benchmarking and Generating Radiology Reports

Ahmed T. Elboardy (Graduate School of Information Science, University of Hyogo), Ghada Khoriba (Center for Informatics Science, School of Information Technology and Computer Science, Nile University), Essam A. Rashed (Advanced Medical Engineering Research Institute, University of Hyogo)

(P05) Linking Pronouns in Dialogue with Whiteboard Symbols: Towards Confusion Detection in Math Collaboration

Chen-Yu Hu (Institute of Science Tokyo), Takuto Asakura (National Institute of Informatics), Koichiro Yoshino (Institute of Science Tokyo)

(P06) Beyond Categories: Learning Continuous Phonological Embeddings for Japanese Sign Language via PU-AUC Optimization

Jundai Inoue (Toyota Technological Institute), Daisuke Hara (Toyota Technological Institute), Makoto Miwa (Toyota Technological Institute)

(P07) Factorizing Relational Data by Many-Body Approximation for Tensors

Takeru Isobe (The Graduate University for Advanced Studies), Katsumi Inoue (National Institute of Informatics), Mahito Sugiyama (The Graduate University for Advanced Studies)

(P08) Robust Gait Recognition in Unseen Environments through Diffusion-Model-Based Data Augmentation

Shinichi Ka (Institute of Science Tokyo), Koichi Shinoda (Institute of Science Tokyo)

(P09) Toward Quantifying Continuous Lexical Semantic Shifts

Hajime Kiyama (Hitotsubashi University), Taichi Aida (Tokyo Metropolitan University), Mamoru Komachi (Hitotsubashi University), Toshinobu Ogiso (National Institute for Japanese Language and Linguistics), Hiroya Takamura (National Institute of Advanced Industrial Science and Technology)

(P10) Spectral Sensitivity Estimation with an Uncalibrated Diffraction Grating

Lilika Makabe (The University of Osaka), Hiroaki Santo (The University of Osaka), Fumio Okura (The University of Osaka), Michael S. Brown (York University), Yasuyuki Matsushita (The University of Osaka)

(P11) Generating Open-Domain Live Commentary with Large Vision-Language Models

Edison Marrese-Taylor (AIST), Erica K. Shimomoto (AIST), Icihro Kobayashi (Ochanomizu University), Yusuke Miyao (The University of Tokyo), Hiroya Takamura (AIST)

(P12) Learning Group Activity Features Through Person Attribute Prediction

Chihiro Nakatani (Toyota Technological Institute Japan), Hiroaki Kawashima (University of Hyogo), Norimichi Ukita (Toyota Technological Institute Japan)

(P13) How Telops Influence Video LLMs

Souto Ohira (Hitotsubashi University), Tosho Hirasawa (Hitotsubashi University), Mamoru Komachi (Hitotsubashi University)

(P14) Age Prediction of Komatsuna using Hu Moments with Neural Networks for Small Datasets

Moeri Okuda (University of Hyogo), Shinsaku Hiura (University of Hyogo)

(P15) Modeling Turn-Taking Speed and Speaker Characteristics

Kazuyo Onishi (Nara Institute of Science and Technology, RIKEN Guardian Robot Project), Hien Onaka (Nara Institute of Science and Technology, RIKEN Guardian Robot Project), Koichiro Yoshino (Nara Institute of Science and Technology, RIKEN Guardian Robot Project, Institute of Science Tokyo)

(P16) Measure Twice, Cut Once: A Semantic-Oriented Approach to Video Temporal Localization with Video LLMs

Zongshang Pang (The University of Osaka), Mayu Otani (CyberAgent, Inc.), Yuta Nakashima (The University of Osaka)

(P17) Integrating LLMs and Supervised Models for Comprehensive Property Extraction from Materials Science Text and Tables

Van-Thuy Phi (RIKEN AIP), Yuji Matsumoto (RIKEN AIP)

(P18) A Foundation Model for Learning Propositional Logic Programs

Yin Jun Phua (Institute of Science Tokyo)

(P19) Data Leakage in Visual Datasets

Patrick Ramos (equal contribution) (The University of Osaka), Ryan Ramos (equal contribution) (The University of Osaka), Noa Garcia (The University of Osaka)

(P20) Processing and acquisition traces in visual encoders: What does CLIP know about your camera?

Ryan Ramos (equal contribution) (The University of Osaka), Vladan Stojnić (equal contribution) (VRG, FEE, Czech Technical University in Prague), Giorgos Kordopatis-Zilos (VRG, FEE, Czech Technical University in Prague), Yuta Nakashima (The University of Osaka), Giorgos Tolias (VRG, FEE, Czech Technical University in Prague)

(P21) Multilingual Evaluation of Large Language Models on the Wordle Word Guessing Game

Matiss Rikters (AIST)

(P22) Sketch2Diagram: Generating Vector Diagrams from Hand-Drawn Sketches

Itsumi Saito (Tohoku University), Haruto Yoshida (Tohoku University), Keisuke Sakaguchi (Tohoku University)

(P23) Rethinking Psychometric Evaluation for LLMs

Jivnesh Sandhan (Kyoto University), Fei Cheng (Kyoto University), Tushar Sandhan (IIT Kanpur, India), Yugo Murawaki (Kyoto University)

(P24) Taming the Untamed: Graph-Based Knowledge Retrieval and Reasoning for MLLMs to Conquer the Unknown

Bowen Wang (The University of Osaka)

(P25) EMMA: Concept Erasure Benchmark with Comprehensive Semantic Metrics and Diverse Categories

Lu Wei (The University of Osaka), Yuta Nakashima (The University of Osaka), Noa Garcia (The University of Osaka)

(P26) Revealing the Impacts of In-Context Learning on Gender Bias in Large Vision-Language Models

Tong Xiang (The University of Osaka), Yuta Nakashima (The University of Osaka), Noa Garcia (The University of Osaka)

(P27) NeuraLeaf: Neural Parametric Leaf Models with Shape and Deformation Disentanglement

Yang Yang (The University of Osaka), Dongni Mao (The University of Osaka), Hiroaki Santo (The University of Osaka), Yasuyuki Matsushita (Microsoft Reseach Tokyo), Fumio Okura (The University of Osaka)

(P28) Hierarchical Group-aware Token Merging for Efficient Trajectory Prediction

Yuki Yoshida (Toyota Technological Institute), Norimichi Ukita (Toyota Technological Institute), Hiromu Taketsugu (Toyota Technological Institute)

(P29) CALICO: Confident Active Learning with Integrated Calibrationn

Lorenzo Querol (The University of Osaka), Hajime Nagahara (The University of Osaka), Hideaki Hayashi (The University of Osaka)