Retrieval context has a sweet spot at Top-K = 10: raising it to 15 did not improve recall but significantly worsened context precision via the distraction issue¶
Claim · All claims
Evidence · 1 study · causalr1·q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly
Subclaims¶
Reading the evidence codes
q is the design tier, i the impact, n the sample, and the kind with its r
the rigour judged within that kind: what each code means.
This claim: no standardized effect is recorded in observations/ for it yet, so no benchmark reading is possible.
q2 i? Increasing retrieval context from Top-K = 10 to 15 did not improve recall but significantly worsened context precision due to irrelevant information distracting the generator. → Anna Ostrowska 2026
Evidence¶
Anna Ostrowska 2026¶
Anna Ostrowska, Michał Kukla, Gabriela Majstrak, Jan Opala, Sebastian Pergała, Jan Skwarek, Anna Wróblewska. (2026). From Surface Learning to Deep Understanding: A Grounded AI Tutoring System for Moodle. https://tinyurl.com/mrytc2dr doi:10.24963/ijcai.2026/984
q2 · i? · causal · r1
Retrieval-setting experiments within the RAGAS evaluation of the system's pipelines. The authors report the "sweet spot" at Top-K = 10 and that raising it to 15 "significantly worsened context precision" without recall gains.
"the ”sweet spot” for context was found to be Top-K = 10; increasing this to 15 did not improve recall but significantly worsened context precision due to the ”distraction issue”"
Discussion¶
Related Claims¶
- Interesting but irrelevant details can impair comprehension and transfer, with effects that vary with the material and the learner — a broader claim this one bears on
- Adding relevant pictures or graphics to text improves learning from it, with a small pooled effect on reading comprehension, though whether dual coding explains the benefit is disputed — related
- A WestEd research bulletin reports a curriculum revision that replaced a decorative photograph in a math problem with a problem-relevant sketch, on the developers' view that such photographs distract students — related
- ChatGPT zero-shot relevance extraction for instruction-quality assessment retrieves mostly irrelevant utterances and overestimates instruction quality — related
- NCERT-tailored RAG preprocessing and figure-aware indexing improve retrieval precision over a generic-chunking baseline, per the authors' assertion — related
- The teacher-conference problem met the sweet spot of complexity: 10th graders generated solution ideas containing the targeted errors without solving the problem canonically — related
Page metadata
| Field | Value |
|---|---|
type |
claim |
id |
rag-top-k-10-sweet-spot-context-precision |
title |
Retrieval context has a sweet spot at Top-K = 10: raising it to 15 did not improve recall but significantly worsened context precision via the distraction issue |
description |
Retrieval context has a sweet spot at Top-K = 10: raising it to 15 did not improve recall but significantly worsened context precision via the distraction issue |
status |
review |
evidence_strength |
moderate |
newest_evidence |
2026 |
recheck_by |
2029 |
generated |
by process:wiki-ingest on 2026-10-11 |
trust |
unverified — no one has confirmed this page's content |
sources |
1 — anna-ostrowska-2026 |
Frontmatter is the page's machine-readable record — see the Schema & Guide.