Bio

I am a fourth-year undergraduate student in Software Engineering at Sichuan University (China). My previous research focused on efficient diffusion models and long video generation.

I am sincerely seeking PhD opportunities for Fall 2027.

I am interested in using visual intelligence to create interactive worlds for exploration and learning. During my PhD, I would like to explore how world models, video generation, LLMs, agents, embodied intelligence, and recursive self-improvement could contribute to this vision.

Publications

BRACE: Taming Sharp Irregularities via Barycentric Rational Forecasting for Fast DiT Inference

JinLong Yang et al. · First author

ACM Multimedia (ACM MM), 2026

A training-free acceleration framework that combines local trajectory modeling, barycentric rational extrapolation, and adaptive Chebyshev weighting for efficient DiT inference.

FADE: Fractional Anomalous Dynamics Extrapolation for Diffusion Transformers Acceleration

JinLong Yang et al. · First author

NeurIPS 2026

Models DiT latent trajectories with fractional dynamics and uses a lightweight lookup strategy for adaptive, zero-overhead extrapolation.

Under Review

LaBACK: Laurent-Guided Backtracking for Training-Free Flow Matching Acceleration Under review

JinLong Yang et al. · First author

A training-free framework that contracts cache-driven updates and applies Laurent-guided forecasting with negative-power correction, delivering consistent speed–quality gains for text-to-image, distilled, and text-to-video generation without extra model evaluations.

HDS: Heterogeneity-Driven Safeguarding for Structure-Aware DiT Acceleration Under review

JinLong Yang et al. · Second author

Profiles inter-layer sensitivity online to safeguard vulnerable layers while accelerating robust ones in Diffusion Transformers.

ExProp: Exponential Propagation on Implicit Manifolds for Training-Free DiT Acceleration Under review

JinLong Yang et al. · Second author

A training-free DiT acceleration method that predicts hidden features through exponential propagation aligned with local dynamics, using a local trust region to constrain propagation.

I keep a reading list.