Fetching from the wire…
Skills2026-03-18 · source-backed
Prepend a 50-100 token LLM-generated explanation to each chunk before embedding and BM25 indexing. Combined contextual embeddings + BM25 cuts top-20 retrieval failure from 5.7% to 2.9%. Add a reranker for 67% reduction. The context generation step is cacheable. Anthropic
Each link below shares sources, entities, or timing with this story.
Anthropic released MCP / Shared entities / Shared topic / What happened next / Tension
Linked by a graph relationship (Anthropic released MCP); both cover Anthropic, BM25, LLM; overlapping topics (context, each).
Anthropic released Mythos / Shared entities / Same source domain / Shared topic / What happened next
Linked by a graph relationship (Anthropic released Mythos); both cover Anthropic, LLM; reported by the same outlet (anthropic.com).
Anthropic released Contextual Retrieval / Shared entities / Shared topic / What happened next
Linked by a graph relationship (Anthropic released Contextual Retrieval); both cover Anthropic, BM25, Combined, LLM; overlapping topics (bm25, chunk, contextual, cuts, embedding).
Anthropic released Claude / Shared entity: Anthropic / Same source domain / Shared topic / What happened next / Downstream implication
Linked by a graph relationship (Anthropic released Claude); both cover Anthropic; reported by the same outlet (anthropic.com).
LLM uses OpenAI / Shared entities / Shared topic / What happened next
Linked by a graph relationship (LLM uses OpenAI); both cover Anthropic, Combined; overlapping topics (context, fewer).
LLM uses OpenAI / Shared entities / Same source domain / Earlier coverage
Linked by a graph relationship (LLM uses OpenAI); both cover Anthropic, LLM; reported by the same outlet (anthropic.com).
Simon Willison released LLM / Shared entities / Shared topic / Earlier coverage
Linked by a graph relationship (Simon Willison released LLM); both cover Anthropic, Combined; overlapping topics (combined, context).
LLM uses OpenAI / Shared entities / Shared topic / Earlier coverage
Linked by a graph relationship (LLM uses OpenAI); both cover Anthropic, LLM; overlapping topics (context, each).