arXiv subjects

Hongyi Ke

Publications and source records attributed to Hongyi Ke.

3 recordsLinked to original sources

arXiv2026

LLM2Manim: Pedagogy-Aware AI Generation of STEM Animations

High-quality STEM animations can be useful for learning, but they are still not common in daily teaching, mostly because they take time and special skills to make. In this paper, we present a semi-automated, human-in-the-loop (HITL) pipeline that uses a large language model (LLM) to help convert math and physics concepts into narrated animations with the Python library Manim. The pipeline also tries to follow multimedia learning ideas like segmentation, signaling, and dual coding, so the narration and the visuals are more aligned. To keep the outputs stable, we use constrained prompt templates, a symbol ledger to keep symbols consistent, and we regenerate only the parts that have errors. We also include expert review before the final rendering, because sometimes the generated code or explanation is not fully correct. We tested the approach with 100 undergraduate students in a within-subject A-B study. Each student learned two similar STEM topics, one with the LLM-generated animations and one with PowerPoint slides. In general, the animation-based instruction gives slightly better post-test scores (83% vs.78%, p < .001), and students show higher learning gains (d=0.67). They also report higher engagement (d=0.94) and lower cognitive load (d=0.41). Students finished the tasks faster, and many of them said they prefer the animated format. Overall, these results suggest LLM-assisted animation can make STEM content creation easier, and it may be a practical option for more classrooms.

cs.MM↗

arXiv2026

Preconditioned Adjoint Data Assimilation for Two-Dimensional Decaying and Forced Turbulence

Adjoint-based data assimilation for turbulent Navier-Stokes flows is limited by backward adjoint growth and increasing dominance of small-scale structures, which degrade reconstruction of initial conditions from sparse measurements. We show that the relative weighting of spectral components can be systematically controlled by redefining the inner product under which the adjoint operator is defined. The resulting Fourier-space weighting kernel acts as a preconditioner for the optimization. Specific kernels correspond to fractional integration or diffusion operators on the initial condition. Numerical experiments show that flow-dependent kernel selection substantially improves reconstruction stability and accuracy: exponential kernels suppress high-wavenumber contributions, whereas a fractional integral kernel is particularly effective for forced Kolmogorov flow. Ensemble statistics of adjoint fields reveal scale-dependent backward growth rates, explaining the instability of the standard formulation and how spectral preconditioning attenuates incoherent small-scale amplification.

physics.flu-dyn↗

arXiv2025

Fourier-Invertible Neural Encoder (FINE) for Homogeneous Flows

We present the Fourier-Invertible Neural Encoder (FINE), a compact and interpretable architecture for dimension reduction in translation-equivariant datasets. FINE integrates reversible filters and monotonic activation functions with a Fourier truncation bottleneck, achieving information-preserving compression that respects translational symmetry. This design offers a new perspective on symmetry-aware learning, linking spectral truncation to group-equivariant representations. The proposed FINE architecture is tested on one-dimensional nonlinear wave interaction, one-dimensional Kuramoto-Sivashinsky turbulence dataset, and a two-dimensional turbulence dataset. FINE achieves an overall 4.9-9.1 times lower reconstruction error than convolutional autoencoders while using only 13-21% of their parameters. The results highlight FINE's effectiveness in representing complex physical systems with minimal dimension in the latent space. The proposed framework provides a principled framework for interpretable, low-parameter, and symmetry-preserving dimensional reduction, bridging the gap between Fourier representations and modern neural architectures for scientific and physics-informed learning.

cs.LG↗