Selected work

Representative Papers

The Last Mile to Production Readiness: Physics-Based Motion Refinement for Video-Based Capture

Dec 14, 2025Proceedings of the SIGGRAPH Asia 2025 Technical Communications

This work addresses the longstanding “last mile” problem in video-based motion capture, where reconstructed motions often exhibit physical implausibility and artifacts that necessitate extensive manual correction for industrial applications such as film and gaming. The authors propose a production-oriented physics-aware motion refinement framework that enhances both single- and multi-person motion sequences through physics-based optimization while seamlessly integrating keyframe editing to allow animators to inject stylistic adjustments. Developed in close collaboration with professional animators, the method balances automation efficiency with artist control, significantly improving the physical plausibility and visual quality of motion data. This approach markedly reduces post-processing effort and is designed to fit directly into real-world animation pipelines.

1 citationsRead paper

RA-ClipScore: Making Generative Model Evaluation More Interpretable

Aug 12, 2026

This work addresses the limited interpretability of existing evaluation metrics for generative models, which struggle to diagnose generation biases at both semantic attribute and spatial distribution levels. To this end, the paper proposes RA-CLIPScore, the first CLIP-based evaluation framework extended to incorporate spatial alignment. RA-CLIPScore employs a dual-prompt mechanism to disentangle competing attributes and leverages local image patch tokens to model region-wise semantic alignment. By introducing a region-specific single-attribute divergence metric, the method substantially enhances interpretability and aligns more closely with human perception. Experiments demonstrate that RA-CLIPScore exhibits greater robustness under distribution shifts or when textual prompts contain partially irrelevant attributes, and its scores show strong correlation with human judgments of visual diversity.

0 citationsRead paper

Capturing Uncertainty in Human Motion for Representation Learning in Soccer

Aug 11, 2026

This work addresses the challenge of capturing the inherent uncertainty in human motion within football scenarios through 3D skeletal representations. To this end, the authors propose a self-supervised representation learning framework that leverages future motion prediction as a proxy task. The approach innovatively introduces a conditional module in 3D Euclidean space to model the multimodal probability distribution of discretized future motions, thereby explicitly capturing multiple plausible motion trajectories. Experimental results demonstrate that the learned representations significantly improve prediction accuracy on large-scale football tracking data and exhibit strong cross-task generalization capabilities across diverse downstream tasks.

0 citationsRead paper

AutoWorldModel-Bench: A State-Centric Benchmark for Automated World-Model Research

Jul 20, 2026

This study addresses the absence of established research paradigms for world models and the inadequacy of existing benchmarks in evaluating open-ended scientific inquiry capabilities. We propose the first automated closed-loop benchmark specifically designed for world model research. By employing unified structured state representations to decouple perception from dynamics modeling, our framework enables coding agents to autonomously iterate and optimize models under fixed computational budgets. Experimental results demonstrate that 94% of sessions achieved performance improvements, with half yielding significant gains; notably, 91% of superior edits constituted substantive refinements to model architectures or training protocols. This work effectively validates agents' capacity for open-ended research and establishes a standardized evaluation framework for advancing world model development.

0 citationsRead paper

Reward-Adaptive Iterative Discovery: A Case Study on Automated Game Testing for NHL26

Jul 08, 2026

This work addresses the reliance on manual, repetitive testing to uncover behavioral vulnerabilities in goalkeeper AI within game development by proposing a multi-agent automated testing framework based on iterative reinforcement learning. The approach incorporates a reward adaptation mechanism into existing reinforcement learning algorithms, effectively mitigating policy overfitting and enabling the continuous discovery of diverse, high-quality vulnerability-inducing strategies. Experimental results demonstrate that a single run of the framework automatically identifies six distinct vulnerability strategies comparable in quality to those found through hours of manual testing, significantly enhancing both testing efficiency and coverage diversity.

0 citationsRead paper

Recent publications

Latest Papers

RA-ClipScore: Making Generative Model Evaluation More Interpretable

Aug 12, 2026

This work addresses the limited interpretability of existing evaluation metrics for generative models, which struggle to diagnose generation biases at both semantic attribute and spatial distribution levels. To this end, the paper proposes RA-CLIPScore, the first CLIP-based evaluation framework extended to incorporate spatial alignment. RA-CLIPScore employs a dual-prompt mechanism to disentangle competing attributes and leverages local image patch tokens to model region-wise semantic alignment. By introducing a region-specific single-attribute divergence metric, the method substantially enhances interpretability and aligns more closely with human perception. Experiments demonstrate that RA-CLIPScore exhibits greater robustness under distribution shifts or when textual prompts contain partially irrelevant attributes, and its scores show strong correlation with human judgments of visual diversity.

0 citationsRead paper

Capturing Uncertainty in Human Motion for Representation Learning in Soccer

Aug 11, 2026

This work addresses the challenge of capturing the inherent uncertainty in human motion within football scenarios through 3D skeletal representations. To this end, the authors propose a self-supervised representation learning framework that leverages future motion prediction as a proxy task. The approach innovatively introduces a conditional module in 3D Euclidean space to model the multimodal probability distribution of discretized future motions, thereby explicitly capturing multiple plausible motion trajectories. Experimental results demonstrate that the learned representations significantly improve prediction accuracy on large-scale football tracking data and exhibit strong cross-task generalization capabilities across diverse downstream tasks.

0 citationsRead paper

AutoWorldModel-Bench: A State-Centric Benchmark for Automated World-Model Research

Jul 20, 2026

This study addresses the absence of established research paradigms for world models and the inadequacy of existing benchmarks in evaluating open-ended scientific inquiry capabilities. We propose the first automated closed-loop benchmark specifically designed for world model research. By employing unified structured state representations to decouple perception from dynamics modeling, our framework enables coding agents to autonomously iterate and optimize models under fixed computational budgets. Experimental results demonstrate that 94% of sessions achieved performance improvements, with half yielding significant gains; notably, 91% of superior edits constituted substantive refinements to model architectures or training protocols. This work effectively validates agents' capacity for open-ended research and establishes a standardized evaluation framework for advancing world model development.

0 citationsRead paper

Reward-Adaptive Iterative Discovery: A Case Study on Automated Game Testing for NHL26

Jul 08, 2026

This work addresses the reliance on manual, repetitive testing to uncover behavioral vulnerabilities in goalkeeper AI within game development by proposing a multi-agent automated testing framework based on iterative reinforcement learning. The approach incorporates a reward adaptation mechanism into existing reinforcement learning algorithms, effectively mitigating policy overfitting and enabling the continuous discovery of diverse, high-quality vulnerability-inducing strategies. Experimental results demonstrate that a single run of the framework automatically identifies six distinct vulnerability strategies comparable in quality to those found through hours of manual testing, significantly enhancing both testing efficiency and coverage diversity.

0 citationsRead paper

Hierarchical Control in Multi-Agent Games: LLM-based Planning and RL Execution

Jun 18, 2026

This work addresses performance bottlenecks in multi-agent reinforcement learning (MARL) arising from sparse rewards, high-dimensional state-action spaces, and the challenge of coordinated policy learning. The authors propose a hierarchical architecture wherein a pretrained large language model (LLM) serves as a centralized strategic controller at the high level, dynamically selecting among specialized low-level reinforcement learning policies without relying on handcrafted rules. This approach represents the first integration of LLMs into high-level planning for multi-agent systems, significantly enhancing tactical diversity and behavioral human-likeness. Evaluated on a 2v2 capture-the-flag task, the method achieves a win rate of 46.4%, matching the performance of hand-designed behavior trees and substantially outperforming flat RL baselines. A user study further reveals that 60% of participants judged the agents’ behavior as most human-like (p = 0.027).

0 citationsRead paper