Research map / Probabilistic and scientific modeling / Dynamical systems
Generalized Correspondence-LDA Models (GC-LDA) for Identifying Functional Regions in the Brain
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Set gamma > 0 (non-zero correspondence smoothing) instead of gamma = 0 (ablation) | GC-LDA (all three variants) · Neurosynth held-out data | held-out log-likelihood (activation + word tokens) | Helped |
| Increase gamma from 0.01 to larger values for activation-only prediction (ablation) | GC-LDA (all three variants) · Neurosynth held-out data | held-out log-likelihood (activation tokens only) | Helped |
| Set gamma in the 0.01 to 0.1 range instead of very small or very large values (ablation) | GC-LDA (all three variants) · Neurosynth held-out data | held-out log-likelihood (word tokens and overall) | Mixed |
| Replace single multivariate Gaussian spatial distribution with a mixture of two Gaussians (unconstrained subregions) (ablation) | GC-LDA spatial representation · Neurosynth held-out data | held-out total log-likelihood | Helped |
| Replace single Gaussian with two symmetric constrained Gaussian subregions (ablation) | GC-LDA spatial representation · Neurosynth held-out data | held-out word-token log-likelihood | Hurt |
| Constrain subregion means to be symmetric about the midline (constrained vs unconstrained subregions) (ablation) | GC-LDA subregions model · Neurosynth held-out data | held-out activation-token log-likelihood | Helped |
| Constrain subregion means to be symmetric about the midline (constrained vs unconstrained subregions) (ablation) | GC-LDA subregions model · Neurosynth held-out data | held-out word-token log-likelihood | Hurt |
| Constrain subregion means to be symmetric about the midline (constrained vs unconstrained subregions) (ablation) | GC-LDA subregions model · Neurosynth held-out data | held-out total log-likelihood | No clear effect |
| Use the constrained subregions model to get interpretable hemispheric lateralization weights | GC-LDA constrained subregions · Neurosynth topics (qualitative) | qualitative recovery of known lateralization | Helped |
Closest papers by meaning
- TopoLM: brain-like spatio-functional organization in a topographic language model (ICLR 2025 · similarity 0.84)
- Neuro-Vision to Language: Enhancing Brain Recording-based Visual Reconstruction and Language Interaction (NeurIPS 2024 · similarity 0.83)
- MindEye2: Shared-Subject Models Enable fMRI-To-Image With 1 Hour of Data (ICML 2024 · similarity 0.83)
- fMRI predictors based on language models of increasing complexity recover brain left lateralization (NeurIPS 2024 · similarity 0.83)
- Finding Shared Decodable Concepts and their Negations in the Brain (ICLR 2025 · similarity 0.83)
- Brain-OF: An Omnifunctional Foundation Model for fMRI, EEG and MEG (NeurIPS 2026 · similarity 0.83)
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.