Unverified50% confidenceFactExact time
Structure-aware chunking uses headings and sections to preserve context; writing the heading path into the text that is embedded helps retrieval matching, and putting it into the context sent to the answer model helps the model understand semantics—these are different stages and cannot substitute for each other.
1
Sources
50%
Confidence
Long-term
Relevance
10/17/2026
First Seen
Sources
Related Entities
Related Claims
Unverified结构化的提示词相比纯文本段落，能让模型更精确地识别不同指令模块的边界，减少指令混淆74% similarUnverifiedChoosing a chunking strategy should start from the types of information the document contains and the questions users will ask: use structure-aware splitting when headings are clear, recursive splitting when formatting is messy, consider semantic chunking when topic shifts are unmarked, use overlap to handle adjacent boundaries, use parent–child retrieval to recover surrounding explanations, and use late chunking to address context loss during embedding; these methods can be combined.74% similarUnverifiedThe contextual retrieval technique involves prepending the document title and section path to each chunk before embedding, or using an LLM to generate a brief summary describing the chunk's position within the document.73% similarUnverified将文本块存储在知识图谱中后，可先检索相关文本块再沿图的关系遍历找到其他相关文本块，为大模型提供更完整上下文73% similarUnverifiedPlacing key instructions in prominent positions, using clear separators to distinguish content types, and annotating retrieved materials with their sources can help the model understand and use context more accurately.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1056904API
curl https://kongchang.com/api/v1/knowledge/claims/1056904MCP
get_claim(id=1056904)