Menu
NEURRATOR (神经叙述器) - 从单个神经元活动生成自然语言描述的框架，实现单细胞分辨率视觉语义叙述
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/hiyenwong/ai_collection --skill neurrate-neural-semantic-narrationThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Based on SOC occupation classification
Showing SKILL.md
| name | neurrate-neural-semantic-narration |
| description | NEURRATOR (神经叙述器) - 从单个神经元活动生成自然语言描述的框架，实现单细胞分辨率视觉语义叙述 |
| version | 1.0.0 |
| category | neuroscience |
| tags | ["neural-encoding","single-neuron","natural-language","clip","multimodal","mouse-visual-cortex"] |
| arxiv | 2606.18667 |
| activation_words | ["神经元叙述","neurator","单细胞分辨率","神经编码","视觉语义","CLIP嵌入","Neuropixel"] |
从单个神经元活动生成自由形式自然语言叙述 - 开创性框架将神经脉冲活动解码为对观看场景的自然语言描述，实现单神经元分辨率级别的语义表征理解。
# 核心架构
class NEURRATOR:
def __init__(self):
self.spike_encoder = SpikeToCLIPEncoder() # 脉冲→CLIP嵌入
self.multimodal_llm = FrozenCLIPLLM() # 多模态语言模型
self.sae_validator = SparseAutoencoder() # SAE验证
def narrate(self, spike_trains, neuron_subset):
# 1. 编码脉冲到CLIP空间
clip_embeddings = self.spike_encoder(spike_trains, neuron_subset)
# 2. LLM生成描述
description = self.multimodal_llm.generate(clip_embeddings)
# 3. SAE验证语义一致性
validated = self.sae_validator.validate(description)
return validated
"神经叙述" (Neurration) - 将细胞身份从分类目标转变为视觉系统的功能性探针，提供神经系统中生物学见解的新单位。