AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

BESPOKE: Benchmark for Search-Augmented Large Language Model Personalization via Diagnostic Feedback

Hyunseo Kim, Sangam Lee, Kwangwook Seo, Dongha Lee

ICML 2026 · Large language models › Retrieval and knowledge · approach: Long-context and personalized retrieval · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add user-history profile (selected, query-aware) (ablation)o3-search · BESPOKEaverage score: 55.93 -> 62.48Improved
add user-history profile (selected, query-aware) (ablation)GPT-4o-search · BESPOKEaverage score: 47.56 -> 57.00Improved
make user context query-aware instead of static (ablation)pplx-sonar · BESPOKEaverage score: 49.76 -> 57.77Improved
select only query-relevant history sessions (ablation)pplx-sonar · BESPOKEaverage score: 57.77 -> 59.74 (with raw vs profile 49.79)Improved
use structured profile instead of raw history (ablation)pplx-sonar · BESPOKEaverage score: 49.79 -> 59.74Improved
add human feedback to the evaluator context (ablation)GPT-5 evaluator · meta-evaluationPearson correlation avg: 0.809 -> 0.847Improved
add personalization context to evaluator (ablation)GPT-5 evaluator · meta-evaluationPearson correlation avg: 0.470 -> 0.809Improved
expand query with CoT before history retrievalStella-V5-1.5B · history retrievalnDCG@10: 0.0820 -> 0.3809Improved
expand query with pseudo-historyStella-V5-1.5B · history retrievalnDCG@10: 0.0820 -> 0.3807Improved
pseudo-history instead of CoT expansionStella-V5-1.5B · history retrievalnDCG@10: 0.3809 vs 0.3807No clear effect
provide gold information instead of web search results (ablation)search-augmented LLMs · BESPOKEneed alignment, content depthImproved
increase search noise to 70% (ablation)pplx-sonar-reasoning · BESPOKE noise studyAvg.: 61.40 -> 49.87Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy