Research map / Vision-language and multimodal / Human and embodied modeling
AVM: Towards Structure-Preserving Neural Response Modeling in the Visual Cortex Across Stimuli and Individuals
Qi Xu, Shuai Gong, Xuming Ran, Haihua Luo, Yangfan Hu
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 changed | Where | Measured | Effect |
|---|---|---|---|
| replace full fine-tuning (V1T-T) with frozen backbone plus modulation paths (AVM) | V1T · Dataset F (adapted from Dataset S) | FEVE: 0.6662 -> 0.7536 | Improved |
| replace training from scratch (V1T-D) with AVM adaptation | V1T · Dataset F (adapted from Dataset S) | FEVE: 0.6363 -> 0.7536 | Improved |
| use frozen backbone with modulation instead of full tuning | V1T · trainable parameters | parameter count: 2.46M -> 0.11M | Improved |
| use block-specific modulation instead of one shared path (AVM vs AVM-S) | AVM-S · Dataset F (adapted from Dataset S) | FEVE: 0.7515 -> 0.7536 | No clear effect |
| add cross-block Cross-CAMU (AVM-B vs AVM) | AVM · Dataset F (adapted from Dataset S) | FEVE: 0.7536 -> 0.7467 | Hurt |
| raise modulation weight from 0.1 to 1.0 (ablation) | AVM · Dataset F (adapted from Dataset S) | FEVE: 0.5507 -> 0.7536 | Improved |
| raise modulation weight from 1.0 to 2.0 (ablation) | AVM · Dataset F (adapted from Dataset S) | FEVE: 0.7536 -> 0.6824 | Hurt |
| raise modulation weight from 1.0 to 2.0 (ablation) | AVM · Dataset S (adapted) | FEVE: 0.6054 -> 0.6094 | Improved |
| increase bottleneck dimension from 1 to 31 (ablation) | AVM · Dataset F (adapted from Dataset S) | FEVE: 0.7219 -> 0.7536 | Improved |
| increase bottleneck dimension from 1 to 100 (ablation) | AVM · Dataset S (adapted) | FEVE: 0.5589 -> 0.6225 | Improved |
| add subject-specific conditioning (AVM vs AVM-S) | AVM-S · Dataset S (mouse C) | FEVE: 0.8271 -> 0.8424 | Improved |
Closest papers by meaning
- NEvo: Neural-Guided Evolutionary Video Synthesis for Dynamic Visual Selectivity (NeurIPS 2026 · similarity 0.86)
- Flexible Context-Driven Sensory Processing in Dynamical Vision Models (NeurIPS 2024 · similarity 0.86)
- Explicitly Modeling Subcortical Vision with a Neuro-Inspired Front-End Improves CNN Robustness (NeurIPS 2025 · similarity 0.86)
- Benchmarking Out-of-Distribution Generalization Capabilities of DNN-based Encoding Models for the Ventral Visual Cortex. (NeurIPS 2024 · similarity 0.85)
- Modeling Dynamic Neural Activity by combining Naturalistic Video Stimuli and Stimulus-independent Latent Factors (NeurIPS 2025 · similarity 0.85)
- DEEM: Diffusion models serve as the eyes of large language models for image perception (ICLR 2025 · similarity 0.85)
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.
