每日文献雷达

2026-07-10

今日入选

• NL2SQLBench: A Modular Benchmarking Framework for LLM-Enabled NL2SQL Solutions（topic keywords matched; recent (≤3 months); code signal; 1 citations）
• Workspace-Bench 1.0: Benchmarking AI Agents on Workspace Tasks with Large-Scale File Dependencies（topic keywords matched; recent (≤3 months); code signal）
• Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs（topic keywords matched; code signal; 1 citations）
• EvoSchema: Towards Text-to-SQL Robustness Against Schema Evolution（code signal; 1 citations）
• Residual Skill Optimization for Text-to-SQL Ensembles（topic keywords matched; recent (≤3 months)）

1. NL2SQLBench: A Modular Benchmarking Framework for LLM-Enabled NL2SQL Solutions

        - **来源**：deepxiv
        - **分数**：0.575
        - **入选原因**：topic keywords matched; recent (≤3 months); code signal; 1 citations
        - **摘要**：Natural Language to SQL (NL2SQL) technology empowers non-expert users to query relational databases without requiring SQL expertise. While large language models (LLMs) have greatly improved NL2SQL algorithms, their rapid development outpaces systematic evaluation, leaving a criti...

• 链接：https://arxiv.org/abs/2604.16493

        <div style="margin-top: 2rem; padding: 1rem; border-left: 3px solid #4a90d9; font-size: 0.9rem; opacity: 0.85;">
        完整深度阅读见博客文章，使用结构化三元组方法拆解论文逻辑。
        </div>
  

2. Workspace-Bench 1.0: Benchmarking AI Agents on Workspace Tasks with Large-Scale File Dependencies

        - **来源**：deepxiv
        - **分数**：0.5675
        - **入选原因**：topic keywords matched; recent (≤3 months); code signal
        - **摘要**：Workspace learning requires AI agents to identify, reason over, exploit, and update explicit and implicit dependencies among heterogeneous files in a worker's workspace, enabling them to complete both routine and advanced tasks effectively. Despite its importance, existing releva...

• 链接：https://arxiv.org/abs/2605.03596

        <div style="margin-top: 2rem; padding: 1rem; border-left: 3px solid #4a90d9; font-size: 0.9rem; opacity: 0.85;">
        完整深度阅读见博客文章，使用结构化三元组方法拆解论文逻辑。
        </div>
  

3. Can LLMs Clean Up Your Mess? A Survey of Application-Ready Data Preparation with LLMs

        - **来源**：deepxiv
        - **分数**：0.53
        - **入选原因**：topic keywords matched; code signal; 1 citations
        - **摘要**：Data preparation aims to denoise raw datasets, uncover cross-dataset relationships, and extract valuable insights from them, which is essential for a wide range of data-centric applications. Driven by (i) rising demands for application-ready data (e.g., for analytics, visualizati...

• 链接：https://arxiv.org/abs/2601.17058

        <div style="margin-top: 2rem; padding: 1rem; border-left: 3px solid #4a90d9; font-size: 0.9rem; opacity: 0.85;">
        完整深度阅读见博客文章，使用结构化三元组方法拆解论文逻辑。
        </div>
  

4. EvoSchema: Towards Text-to-SQL Robustness Against Schema Evolution

        - **来源**：deepxiv
        - **分数**：0.4925
        - **入选原因**：code signal; 1 citations
        - **摘要**：Neural text-to-SQL models, which translate natural language questions (NLQs) into SQL queries given a database schema, have achieved remarkable performance. However, database schemas frequently evolve to meet new requirements. Such schema evolution often leads to performance degr...

• 链接：https://arxiv.org/abs/2603.10697

        <div style="margin-top: 2rem; padding: 1rem; border-left: 3px solid #4a90d9; font-size: 0.9rem; opacity: 0.85;">
        完整深度阅读见博客文章，使用结构化三元组方法拆解论文逻辑。
        </div>
  

5. Residual Skill Optimization for Text-to-SQL Ensembles

        - **来源**：deepxiv
        - **分数**：0.4875
        - **入选原因**：topic keywords matched; recent (≤3 months)
        - **摘要**：Text-to-SQL ensembles improve over single-candidate generation by drawing multiple SQL candidates and selecting one, but their effectiveness is bounded by Pass@K, the probability that at least one of K candidates is correct. Existing methods source diversity heuristically through...

• 链接：https://arxiv.org/abs/2605.21792

        <div style="margin-top: 2rem; padding: 1rem; border-left: 3px solid #4a90d9; font-size: 0.9rem; opacity: 0.85;">
        完整深度阅读见博客文章，使用结构化三元组方法拆解论文逻辑。
        </div>