Query refinement transforms an information need into a more effective query or sequence of sub-queries. RQ-RAG makes this behavior an explicit model capability rather than a runtime prompt wrapper. section 7.6.5
RQ-RAG supports four refinement actions:
- Skip retrieval for trivial conversational input. section 7.6.5
- Rewrite an unclear or ineffective query. section 7.6.5
- Decompose a multi-hop question into sequential sub-queries. section 7.6.5
- Disambiguate user intent when the request is unclear. section 7.6.5
Chained multi-hop refinement
The question “What kind of university is the school where Rey Ramsey was educated?” becomes “Where was Rey Ramsey educated?” → “Rutgers University” → “What university is Rutgers University an instance of?” → “Land-grant University.” Partial answers guide later retrieval, while trajectory selection chooses among candidate refinement paths. section 7.6.5
Table 7.2 Comparison of HyDE variants and query-side techniques
|
Variation
|
Primary benefit
|
Implementation cost/complexity
|
Latency impact
|
Use case
|
|
Standard HyDE
|
Significant zero-shot retrieval improvement
|
Low
|
Medium
|
General-purpose RAG where baseline retrieval is poor.
|
|
Multi-Sample HyDE
|
Improved stability and robustness to noise
|
Medium (requires averaging)
|
High
|
Applications where query robustness is critical.
|
|
Hybrid HyDE (Fusion)
|
Combines semantic and lexical strengths
|
High (requires fusion logic)
|
Medium
|
High-precision search that cannot miss keyword matches.
|
|
HyDE + Re-ranker
|
Maximizes precision for top results
|
High (requires a second model)
|
High
|
Safety-critical or expert systems (e.g., legal, medical).
|
|
Reverse HyDE
|
Eliminates query-time LLM latency
|
Medium (shifts cost to indexing)
|
None at query time
|
Read-heavy corpora that change rarely (docs, catalogs,
KBs).
|
|
RQ-RAG
|
Built-in rewrite / decompose / disambiguate; strong on
multi-hop
|
High (requires supervised fine-tuning of the generator)
|
High (multiple sequential refinements per query)
|
Systems where query patterns are complex and a
custom-trained generator is acceptable.
|
Compare query-side retrieval techniques by benefit, implementation complexity, latency, and use case. section 7.6.5

