Phase 5 advanced retrieval

Phase 5 — Risks and Mitigations

Risks for hybrid retrieval, graph traversal, reranking, evaluation, and shadow rollout of advanced retrieval.

Phase 5 — Risks and Mitigations

RiskAreaLikelihoodImpactMitigation
Hybrid retrieval improves some queries while regressing others5.A.xMediumHighCompare dense-only, hybrid, reranked, and graph-aware modes side by side; keep dense-only as a baseline floor in CI
BM25 tuning is corpus-sensitive5.A.1MediumMediumStart with Postgres-native ranking, measure on the gold set, and leave room for later investigation if corpus behavior suggests additional tuning
Cross-encoder reranking adds too much latency5.A.3MediumHighBound reranking to a small candidate set, keep a clear timeout budget, and fall back cleanly to fused ranking when the reranker is unavailable
Graph traversal leaks tenant boundaries or follows pathological cycles5.B.1LowCriticalEnforce org_id filtering at every recursive step, cap traversal depth in SQL, and keep explicit cycle fixtures in tests
Contradiction filtering hides useful context5.B.2MediumMediumTreat the filter as a conservative packing rule, log dropped contradictions for review, and revisit the strategy if evaluation results show missed recall
Gold-set coverage is too narrow for hybrid and graph behaviors5.C.1MediumHighSplit lexical and graph-specific subsets, document label quality, and keep expansion of the corpus as an ongoing research task
Shadow evaluation creates operational noise without decision value5.C.3MediumMediumSample conservatively, store divergence metrics asynchronously, and review dashboard usefulness before raising sampling rates
Retrieval-quality expectations outpace underlying data quality5 overallMediumMediumKeep benchmarking tied to labeled evidence, record open questions in findings/exit audit, and allow follow-up investigation where quality deltas are unclear
Edit on GitHub

Last updated on

On this page

No Headings