Unverified50% confidenceFactExact time
Contextual biasing is a technique that introduces an external vocabulary or phrase list at the decoding stage to increase the recognition probability of specific word sequences by adjusting language model scores or using methods such as shallow fusion.
1
Sources
50%
Confidence
Long-term
Relevance
11/10/2026
First Seen
Sources
Related Entities
Related Claims
Unverified上下文偏置/热词偏置在解码阶段通过WFST或注意力机制动态提升特定词汇的输出概率,实现方式包括浅层融合和深层偏置70% similarUnverifiedContext Engineering focuses on the flow of information across multiple rounds of conversation, unlike prompt engineering, which focuses on the wording of a single input.69% similarVerifiedContext-aware generation requires the model not only to grasp what has changed, but also to express those changes in an appropriate tone and format while maintaining consistency with the existing documentation system.69% similarUnverifiedsemanticfa is an R package that implements semantic factor analysis: it converts scale items into semantic embeddings and performs factor analysis directly on the semantic similarity matrix among items, without requiring respondent response data.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1153742API
curl https://kongchang.com/api/v1/knowledge/claims/1153742MCP
get_claim(id=1153742)