Chi Zhang | PKU CoRe Lab

Chi Zhang

Assistant Professor

Peking University

Latest

[ICLR26] Learning physics-grounded 4D dynamics with neural gaussian force fields
[ICLR26] Neural Force Field: Few-shot Learning of Generalized Physical Reasoning
[CHI26] NarrativeLoom: Enhancing Creative Storytelling through Multi-Persona Collaborative Improvisation
[NatureMachineIntelligence26] Proposing and solving olympiad geometry with guided tree search
[NeurIPS25] Heterogeneous Adversarial Play in Interactive Environments
[CogSci25] Probing and Inducing Combinational Creativity in Vision-Language Models
[CogSci25] A simulation-heuristics dual-process model for intuitive physics
[ScienceAdvances24] Human-level few-shot concept induction through minimax entropy learning
[ICLR24] I-PHYRE: Interactive Physical Reasoning
[NeurIPS23] Active Reasoning in an Open-World Environment
[NeurIPS23] Evaluating and Inducing Personality in Pre-trained Language Models
[NeurIPS23] Interactive Visual Reasoning under Uncertainty
[ICCV23] X-VoE: Measuring eXplanatory Violation of Expectation in Physical Events
[ICML23] MEWL: Few-shot multimodal word learning with referential uncertainty
[NeurIPS22] On the Learning Mechanisms in Physical Reasoning
[ECCV22] Learning Algebraic Representation for Systematic Generalization in Abstract Reasoning
[CVPR21] ACRE: Abstract Causal Reasoning Beyond Covariation
[CVPR21] Abstract Spatial-Temporal Reasoning via Probabilistic Abduction and Execution
[ICRA21] Congestion-aware Multi-agent Trajectory Prediction for Collision Avoidance
[Engineering20] Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense
[AAAI20] Machine Number Sense: A Dataset of Visual Arithmetic Problems for Abstract and Relational Reasoning
[NeurIPS19] Learning Perceptual Inference by Contrasting
[IROS19] Learning Virtual Grasp with Failed Demonstrations via Bayesian Inverse Reinforcement Learning
[CVPR19] RAVEN: A Dataset for Relational and Analogical Visual Reasoning
[AAAI19] MetaStyle: Three-Way Trade-Off Among Speed, Flexibility and Quality in Neural Style Transfer
[AAAI19] Mirroring without Overimitation: Learning Functionally Equivalent Manipulation Actions