cs.CLAug 7, 2026

From Test-Time Scaling to Reusable Memory: Measuring Crystallization in Text-to-SQL

Authors: Jiaqian WangYutao QiWenjin HouYuanxi CheMuning Wen

Organizations: Xidian University · Xidian University Xi’an, Shaanxi, China · Shanghai Jiao Tong University · Shanghai Jiao Tong University School of Artificial Intelligence Shanghai, China

Abstract

Test-time scaling can correct difficult text-to-SQL queries, but the extra computation is normally discarded after each answer. Systems increasingly retain verified repair episodes, yet evaluations still report one end-to-end score. It cannot distinguish replay on recurring questions from help on unseen questions, or identify the responsible memory choice. We call measuring this future value the crystallization problem. Our controlled evaluation holds the single-shot solver fixed and varies one memory choice at a time. We separately measure replay, cross-question retention, and held-out same-database transfer. On BIRD, storing verified corrected queries improves held-out first-attempt accuracy by 4.34 percentage points. This gain captures 44.4% of the accuracy headroom provided by on-demand repair on the same questions. Controlled interventions identify database-specific content as the main operating ingredient. Reliable verification and broader retrieval coverage yield supported gains; richer formats and elaborate retrievers do not. Open-source code, evaluation artifacts, and reproduction instructions are available at https://github.com/ai-jiaqian/text-to-sql-memory-crystallization.

Explore similar work

CardsList
  1. PExA: Parallel Exploration Agent for Complex Text-to-SQL

    Apr 24, 2026Tanmay Parekh, Ella Hofmann-Coyle, Shuyi Wang +3Text-To-SqlSql

  2. R3^3-SQL: Ranking Reward and Resampling for Text-to-SQL

    Apr 28, 2026Hojae Han, Yeonseok Jeong, Seung-won Hwang +2Text-To-SqlSql