Tenth International Workshop on Symbolic-Neural Learning (SNL2026)

June 25-26, 2026
Toyota Technological Institute at Chicago (Chicago, IL, USA)

Program

Thursday, June 25
09:00–09:50	Registration + Light Breakfast (5th floor)
09:50–10:00	Opening Remarks
10:00–11:00	Keynote Talk I	Chenhao Tan	Science in the Age of AI
11:00–11:30	Invited Talk I	Hao Peng	RoPE at the End of Its Rope
11:30–12:00	Invited Talk II	Lu Wang	Building Trustworthy Large Language Models: Methods and Evaluation
12:00–13:15	Lunch (Provided)
13:15–14:15	Keynote Talk II	Gaël Richard	Symbolic Information and Priors for Neural Music Processing
14:15–14:45	Invited Talk III	Michael Maire	Generative Learning of Visual Representations
14:45–15:15	Invited Talk IV	Manling Li	Multimodal Reasoning Intelligence: How Foundation Models Learn (and Fail to Learn) Physical World
15:15–15:30	Break
15:30–16:30	Poster Session (4th floor)
16:30–17:30	Keynote Talk III	Joyce Chai	Language Learning and Grounding through Social and Embodied Experience
18:00–21:00	Reception at The Study
Friday, June 26
09:00–09:30	Light Breakfast
09:30–10:30	Keynote Talk IV	James Evans	The Singularity will not be Singular: Complexity or Collapse in AI & Collective Intelligence
10:30–11:00	Invited Talk V	Yuxin Chen	Learning from Symbolic Feedback: From Textual Gradients to Scientific Discovery
11:00–11:15	Break
11:15–12:15	Keynote Talk V	Mohit Bansal	Controllable, Long-Horizon Planning for Skill Discovery and Multimodal Generation
12:15–12:45	Invited Talk VI	Norimichi Ukita	Human-like Stochastic Motion Generation and Prediction
12:45–12:50	Closing Remarks
12:50–14:00	Lunch (Provided)

Poster Session

(P01) Measuring, Localizing, and Ablating Alignment Signatures in LLMs

Aniket Anand (University of Chicago), Janvijay Singh (University of Illinois Urbana-Champaign), Zhewei Sun (Toyota Technological Institute at Chicago), Dilek Hakkani-Tür (University of Illinois Urbana-Champaign), Nick Feamster (University of Chicago)

(P02) CLAW: Learning Continuous Latent Action World Models End-to-End via Adversarial Latent Regularization

Tewodros Ayalew (University of Chicago), Matthew Jeung (Toyota Technological Institute at Chicago), Xiao Zhang (University of Chicago), Samuel Wheeler (Argonne National Laboratory), Kaylene Caswell Stocking (Toyota Technological Institute at Chicago), Michael Maire (University of Chicago), Matthew Walter (Toyota Technological Institute at Chicago)

(P03) ChiKhaPo: A Large-Scale Multilingual Benchmark for Evaluating Lexical Comprehension and Generation in Large Language Models

Emily Chang (Toyota Technological Institute at Chicago), Niyati Bafna (Johns Hopkins University)

(P04) Scaling Textual Gradients via Sampling-Based Momentum

Yuxin Chen (University of Chicago), Zixin Ding (University of Chicago), Junyuan Hong (National University of Singapore), Zhan Shi (University of Texas at Austin), Jiachen T. Wang (Princeton University), Zinan Lin (Microsoft Research), Li Yin (SylphAI), Meng Liu (SylphAI), Zhangyang Wang (University of Texas at Austin)

(P05) Teaching an Agent to Sketch One Part at a Time

Xiaodan Du (Toyota Technological Institute at Chicago), Ruize Xu (University of Chicago), David Yunis (Toyota Technological Institute at Chicago), Yael Vinker (Massachusetts Institute of Technology), Greg Shakhnarovich (Toyota Technological Institute at Chicago)

(P06) Identifying and Typifying Demographic Unfairness in Phoneme-Level Embeddings of Self-Supervised Speech Recognition Models

Felix Herron (Université Grenoble Alpes), François Portet (Université Grenoble Alpes), Alexandre Allauzen (ESPCI Paris — PSL), Solange Rossato (Université Grenoble Alpes)

(P07) LLM-Driven Induction of Expressive Schemas for Interpretable Knowledge Graph Completion

Ryuki Ida (Toyota Technological Institute), Makoto Miwa (Toyota Technological Institute)

(P08) Enhancing LLMs with Relative Awareness of Physical Properties

Ulin Nuha (Toyota Technological Institute), Kazuma Hayakawa (Toyota Technological Institute), Yutaka Sasaki (Toyota Technological Institute)

(P09) Prior Knowledge Makes It Possible: From Sublinear Graph Algorithms to LLM Test-Time Methods

Donya Saless (Toyota Technological Institute at Chicago), Avrim Blum (Toyota Technological Institute at Chicago), Daniel Hsu (Columbia University), Cyrus Rashtchian (Google Research)

(P10) Flow-based Policy Adaptation without Policy Updates

Luzhe Sun (Toyota Technological Institute at Chicago), Jingtian Ji (Toyota Technological Institute at Chicago), Haoran Chen (Toyota Technological Institute at Chicago), Jiawei Zhou (Stony Brook University), Matthew R. Walter (Toyota Technological Institute at Chicago)

(P11) Hexa-P: Physical Plausibility-aware Pre-training for Pedestrian Path Prediction

Hiromu Taketsugu (Toyota Technological Institute), Shohei Nobuhara (Kyoto Institute of Technology), Norimichi Ukita (Toyota Technological Institute)

(P12) Re-EmLoco: Recovering Embodied Locomotion for Degradation-Robust Pedestrian Path Prediction

Hiromu Taketsugu (Toyota Technological Institute), Shohei Nobuhara (Kyoto Institute of Technology), Norimichi Ukita (Toyota Technological Institute)

(P13) 2D MotionGPT: Toward Motion Language Models for Real-World Applications

Kaname Yokoyama (Toyota Technological Institute), Norimichi Ukita (Toyota Technological Institute)