Unverified50% confidenceFactExact time
RRT（快速随机扩展树）算法由Steven LaValle于1998年提出，通过在配置空间随机采样搜索无碰撞路径，适合高维关节空间的在线规划
1
Sources
50%
Confidence
Long-term
Relevance
7/16/2026
First Seen
Sources
Related Claims
Unverified斯坦福大学与Anyscale合作开发的RouteLLM框架，2024年论文证明基于RLHF偏好数据训练的路由器可在将50%请求路由至GPT-3.5级模型的前提下保持整体任务得分在GPT-4的95%水平以上61% similarUnverifiedTree-of-Thought（ToT）在 CoT 基础上引入树状搜索结构，允许模型同时探索多条推理路径并回溯58% similarUnverifiedGPT-4o和Claude 3.5为代表的新一代模型通过RLHF和大规模工具调用训练，使主动发现问题在工程层面成为可能55% similarVerifiedLlamaIndex (formerly GPT Index) was created by Jerry Liu in 2022, focusing on solving the connection problem between large models and external data.54% similarUnverifiedMainstream LLMs such as the GPT series use a decoder-only autoregressive architecture that generates text by predicting the probability distribution of the next token, with a training objective of minimizing cross-entropy loss.54% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/534535API
curl https://kongchang.com/api/v1/knowledge/claims/534535MCP
get_claim(id=534535)