research(nightly): adaptive-sq-coherence — mixed-precision SQ with coherence-precision routing#702
Draft
ruvnet wants to merge 1 commit into
Draft
Conversation
Topic: Adaptive Scalar Quantization with Coherence-Precision Routing Slug: adaptive-sq-coherence Date: 2026-07-17 ADR: ADR-272 Connects: vector search, coherence scoring, agent memory, edge WASM, RVF. Co-Authored-By: claude-flow <ruv@ruv.net> Claude-Session: https://claude.ai/code/session_01Rgb8BSCix9faJMAPNCaGmJ
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Nightly RuVector research for 2026-07-17: Adaptive Scalar Quantization with Coherence-Precision Routing.
Core idea: route each stored vector to 8-bit or 16-bit scalar quantization at insert time, based on its local neighbourhood density score (mean kNN L2 distance). Dense, contested embedding regions get 16-bit precision; sparse regions get 8-bit. The result is a memory:recall Pareto improvement over both uniform options.
All numbers from
cargo run --release -p ruvector-adaptive-sq --bin benchmark, seed=42, N=5000, dim=32, 200 queries, k=10.Deliverables
crates/ruvector-adaptive-sq— zero external deps (library), builds to WASM unmodifiedcargo test -p ruvector-adaptive-sq)cargo run --release -p ruvector-adaptive-sq --bin benchmarkdocs/adr/ADR-272-adaptive-sq-coherence.mddocs/research/nightly/2026-07-17-adaptive-sq-coherence/README.mddocs/research/nightly/2026-07-17-adaptive-sq-coherence/gist.mdEcosystem Connections
Connects: RuVector vector search + coherence scoring + agent memory + edge WASM + RVF + ruFlo.
Known Limitations (documented in ADR)
Build Status
Generated by Claude Code