Unverified50% confidenceFactTime unknown
BM25 solves TF-IDF's problem of inflated term frequency in long documents by compressing a single term's weight in longer documents
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Related Claims
Unverified即便是支持128K上下文的模型，在文档长度超过32K时，中段信息的检索准确率往往出现明显下滑58% similarUnverifiedWAL文件异常膨胀造成的持续小块随机写入相比顺序大块写入会导致更高的写入放大系数，加速SSD寿命消耗57% similarUnverified只有在LLM推理比embedding加向量检索慢数十倍以上时，语义缓存才能同时实现降本与提速56% similarUnverifiedBERT单条文本推理延迟通常在数十毫秒量级，显存占用较大，在高并发系统中处理能力受限55% similarUnverified2016年，Stack Overflow因一个用于去除行尾空白字符的正则表达式遇到异常长的空格序列而宕机28分钟54% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/55834API
curl https://kongchang.com/api/v1/knowledge/claims/55834MCP
get_claim(id=55834)