| CEVAL | Contrastive Evaluation | 对比评估 | |
| CLIP | Contrastive Language-Image Pre-training | 对比语言图像预训练 | |
| CMMLU | Cross-Modal Multi-Modal Language Understanding | 跨模态多模态语言理解 | |
| FFN | Feed-Forward Network | 前馈网络 | |
| GGML | GPT-Generated Model Language | GPT生成的模型语言 | |
| GGUF | GPT-Generated Unified Format | GPT生成的统一格式 | |
| GQA | Grouped Query Attention | 分组查询注意力 | |
| LaMDA | Language Model for Dialogue Applications | 对话应用的语言模型 | 2021, by Google |
| LLaMA | Large Language Model Meta AI | 大型语言模型Meta AI | 2023, by Meta |
| LLM | Large Language Model | 大型语言模型 | |
| LoRA | Low-Rank Adaptation | 低秩适配 | |
| MHA | Multi-Head Attention | 多头注意力 | |
| MMLU | Multi-Modal Language Understanding | 多模态语言理解 | |
| MQA | Multi-Query Attention | 多查询注意力 | |
| PEFT | Parameter-Efficient Fine-Tuning | 参数高效微调 | |
| ppl | perplexity | 困惑度 | |
| QKV | Query, Key, Value | 查询、键、值 | |
| RAG | Retrieval-Augmented Generation | 检索增强生成 | |
| RLHF | Reinforcement Learning from Human Feedback | 来自人类反馈的强化学习 | |
| RTN | Reversible Transformer Network | 可逆变换网络 | |
| SFT | Self-Supervised Fine-Tuning | 自监督微调 | |
| SFT | Supervised Fine-Tuning | 监督微调 | |
| TTS | Text-to-Speech | 文本到语音 | |
| VLM | Visual Language Model | 视觉语言模型 | |
| VQA | Visual Question Answering | 视觉问答 | |