liveBook · Manning

Contriever

As explained in Retrieval Augmented Generation, The Foundational Ideas

Contriever is a dense retrieval model that learns representations for retrieving relevant documents without supervised retrieval labels.

When a question-answering system needs current information, keeping documents in an external index is often more practical than retraining its language model. Contriever is a dense retriever that represents text as vectors. It learns from unsupervised contrastive training rather than human-labeled query–document pairs, making it useful when relevance labels are unavailable. section 6.1 section 7.3

In Atlas, Contriever supplies the retrieval component. Transformer-based encoders embed queries and documents; document vectors are computed ahead of time and stored in an index. At query time, dot-product similarity ranks documents against the query vector. The retrieved information can change when the index is updated, without changing the language model’s parameters. section 6.1

Searching with a hypothetical answer

HyDE takes a different route to the query vector: it generates a hypothetical answer passage, then embeds that passage with the same encoder used for the real documents. Contriever was the encoder in the original HyDE work. Its fixed-size embedding can discard an incorrect date or name while retaining the passage’s topic. It cannot rescue a passage framed around the wrong subject: retrieval may follow that mistaken framing. section 7.3

Table 7.1 HyDE zero-shot performance vs. baselines
Benchmark
BM25 (Lexical)
Contriever (Unsupervised Dense)
HyDE (Zero-Shot)
Contriever-FT (Supervised Dense)
TREC DL19
38.3
44.5
61.9
64.5
TREC DL20
35.8
43.1
60.1
62.1
BEIR (SciFact)
66.8
64.3
69.7
71.0
BEIR (FiQA)
26.9
30.5
38.8
48.1

The benchmark table places unsupervised Contriever beside lexical BM25, zero-shot HyDE, and supervised fine-tuned Contriever. It shows why adding a hypothetical passage can help, while giving a reason to evaluate retrieval on the intended dataset rather than assume the more elaborate pipeline always wins. section 7.3

How Retrieval Augmented Generation, The Foundational Ideas covers it

How deep it goes
Explains embeddings, unsupervised contrastive training, dot-product matching, and HyDE’s lossy bottleneck, but does not provide a Contriever training recipe.
Examples
Compares BM25, Contriever, HyDE, and fine-tuned Contriever across TREC DL19, TREC DL20, SciFact, and FiQA.
What it assumes
Introduces the embedding and matching mechanism while assuming familiarity with queries, document corpora, and indexes.
What you can do after
Evaluate/understand it as an unsupervised retrieval baseline and identify when HyDE’s hypothetical passage helps or misdirects search.
How it connects
Connects dual encoders and vector indexes to retrieval-augmented generation, unsupervised contrastive learning, and hypothetical document embeddings.

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Retrieval Augmented Generation, The Foundational Ideas. The section links point to the parts of the book it draws on.