1.  [](/)
2.  [Diagrams](/diagrams)
3.  MUVERA: Which Knob Actually Buys Recall (FDE Dimensionality Ablation)

[Back to Diagrams](/diagrams)MUVERA / Multi-Vector

# MUVERA: Which Knob Actually Buys Recall (FDE Dimensionality Ablation)

Which Knob Actually Buys Recall (FDE Dimensionality Ablation)

![Recall@100 versus final FDE dimensionality on nfcorpus with ColBERTv2, log x-axis, one line per swept knob (k\_sim, d\_proj, r\_reps); r\_reps dominates at low dimensionality, d\_proj floors near 16, and k\_sim saturates past 5.](/diagrams/muvera-recall-ablation.webp)Full screen

Which Knob Actually Buys Recall (FDE Dimensionality Ablation)

An FDE's size is the product of three knobs (final dimensionality = B x d\_proj x R\_reps, where B = 2^k\_sim SimHash buckets) and this ablation (nfcorpus, ColBERTv2, Recall@100 vs exact top-10) sweeps each knob on its own from the k\_sim=5, d\_proj=16, r\_reps=20 baseline (10,240d, recall around 0.73). The knobs are not interchangeable. r\_reps (the number of repetitions) dominates at low dimensionality: recall climbs from 0.25 at r=1 to 0.62 by r=8, so early repetitions are where the recall is bought. d\_proj (the per-bucket projection dimension) floors near 16, pushing it to 64 barely moves the line. k\_sim (the number of SimHash hyperplanes) saturates past 5. The practical punchline is in the takeaway strip: because MUVERA always follows tier 1 with an exact rerank, NDCG@10 moves only 0.305 to 0.332 going from 2,560d to 10,240d: a 4x disagreement in FDE size mostly washes out after reranking, so the cheap first-tier index can run far smaller than intuition suggests.

## Where this diagram appears

*   [Guide: Late-Interaction Retrieval (ColBERT, ColPali, MUVERA) →](/guides/late-interaction-retrieval)

Run this on your own data

Mixpeek turns video, images, audio, and documents in your object storage into searchable, timestamped results through one API.

[Search your own data](https://studio.mixpeek.com/namespaces/create?mode=managed&utm_source=homepage&utm_medium=diagram_detail&utm_campaign=mvs_onramp)

## Related diagrams

[

![An embedding turns meaning into coordinates: a model reads 'dog', 'puppy', and 'invoice' and outputs vectors, so 'dog' and 'puppy' land near each other while 'invoice' lands far away. Text, images, audio clips, video frames, and faces all map into one shared coordinate space where distance measures difference in meaning.](/diagrams/foundations/embeddings.webp)

An embedding is meaning turned into coordinates



](/diagrams/embeddings)[

![A variable-size set of token vectors becomes one fixed-size vector via SimHash bucketing, projection, and repetition: no neural encoder required.](/diagrams/muvera-fde-pipeline.webp)

MUVERA: How a Multi-Vector Set Becomes One Fixed Vector



](/diagrams/muvera-fde-pipeline)[

![Documents fill empty buckets with a mean; queries sum and leave empty buckets zero: the asymmetry is what makes the FDE dot product approximate Chamfer / MaxSim.](/diagrams/muvera-query-doc-asymmetry.webp)

MUVERA: Why Query and Document Are Encoded Asymmetrically



](/diagrams/muvera-query-doc-asymmetry)