语义回响：通过回收被丢弃Token嵌入增强语言模型表达能力
-
Updated
Aug 7, 2026 - Python
#
semantic-echo
Here are 3 public repositories matching this topic...
Semantic Echo - A 1.5B LLM beats big-lab baselines on 4/5 Turing tests (emotion/human-likeness). Raw experiment data: 19-config multi-model, thought-chain fallback, full-pipeline, papers, charts. 语义回响 · 1.5B 反超大厂实证数据仓库
nlp machine-learning ai deep-learning sentiment-analysis emotion inference reasoning turing-test experiment-data large-language-models llm chinese-llm semantic-echo fine-tuning-free
-
Updated
Aug 7, 2026 - Python
1.5B小模型反超大厂！语义回响(Semantic Echo)推理架构实证｜不修改权重、不重训练，仅靠回收被丢弃的Token嵌入增强情感表达｜5项图灵测试4/5反超，人似度翻倍
python chatbot emotion transformer chinese language-model human-like turing-test llm human-likeness no-fine-tuning small-model inference-architecture semantic-echo emotion-enhancement emotional-expression inference-enhancement 1-5b beats-big-labs
-
Updated
Aug 7, 2026 - Python
Improve this page
Add a description, image, and links to the semantic-echo topic page so that developers can more easily learn about it.
Add this topic to your repo
To associate your repository with the semantic-echo topic, visit your repo's landing page and select "manage topics."