What happened: Effective Dense Retrieval using Only In-Context Examples

RICE method enables large language models to generate effective dense retrieval embeddings using only in-context examples, without requiring model training.

Why it matters: Effective Dense Retrieval using Only In-Context Examples

Relevant for engineers building search or retrieval systems who want to leverage existing LLMs without the cost and complexity of fine-tuning retrievers.

In the news

Effective Dense Retrieval using Only In-Context Examples

arXiv cs.AI · Published September 29, 2026 · 3 min read

In 30 seconds

What happened: RICE method enables large language models to generate effective dense retrieval embeddings using only in-context examples, without requiring model training.
Why it matters: Relevant for engineers building search or retrieval systems who want to leverage existing LLMs without the cost and complexity of fine-tuning retrievers.
Watch out: Paper is recent preprint; real-world performance compared to trained retrievers and scalability to production systems remain to be validated.

Read the full article

llm
language model
retrieval
retriever
prompt

The patterns behind this

Each one covers how the technique works, when it earns its cost, and where it breaks.

The Agent Architect

One pattern, one tradeoff, one production failure story. A short weekly briefing for people building agentic systems.

Weekly email, one-click unsubscribe. We only use your address to send the briefing.