AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

MOD-SR: Unifying Multimodal Learning and Direct Optimization with Gradient-Guided Diffusion Model for Symbolic Regression

Chuyang Xiang, Yichen Wei, Junchi Yan

ICML 2026 · Image generation › Fast diffusion sampling · approach: Distributional control of diffusion

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add MSE gradient guidance at inference (ablation)MOD-SR · Standard benchmarks averageR2: 0.968 -> 0.976Improved
add length gradient guidance at inference (ablation)MOD-SR · Standard benchmarks averagecomplexity (nodes): 17.0 -> 16.0Improved
add MSE gradient guidance (ablation)MOD-SR · ConstantR2: 0.929 -> 0.983Improved
add MSE gradient guidance (ablation)MOD-SR · Keijzer, Koza, Livermore, RR2: Keijzer 0.970 -> 0.970No clear effect
add representation alignment during training (ablation)MOD-SR · training dynamicsconvergence and final performanceImproved
use random subtree masking instead of full-tree optimization in guidance (ablation)MOD-SR · hard casesfitImproved
use native perturbation with gradient guidance instead of GP or LSO post-processingMOD-SR · Standard benchmarksR2: Keijzer 0.819 (SNIP LSO) -> 0.983Improved
use token-wise embedding space instead of pooled global vector (ablation)SNIP encoder · expression reconstructionaccuracy: 0.6518 -> 0.9997Improved
enable gradient guidance at inference (ablation)MOD-SR · Single-instance runtimeruntime: 2.80s -> 18.44sHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy