Unverified50% confidenceFactExact time
Compact mode compresses and merges multiple retrieved text chunks before sending them to the LLM to reduce the number of API calls, while refine mode iteratively sends each text chunk to the LLM one by one to progressively refine the answer.
1
Sources
50%
Confidence
Long-term
Relevance
10/4/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedLarge models process text using Token (word element) as the minimum processing unit, with tokenizers handling the segmentation.70% similarUnverified轻量模型在多语言覆盖和长文本处理上普遍存在短板69% similarUnverified现代LLM的分词器通常将文本压缩为约10万词汇量的Token序列68% similarUnverifiedAlignment is a core concept in multimodal large models, and its essence is to spatially align non-textual information and convert it into a token sequence that an LLM can understand.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/971996API
curl https://kongchang.com/api/v1/knowledge/claims/971996MCP
get_claim(id=971996)