Research map / Large language models / Language, safety and interpretability
FANDA: A Novel Approach to Perform Follow-Up Query Analysis
Qian Liu, Bei Chen, Jian-Guang Lou, Ge Jin, Dongmei Zhang
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 changed | Where | Measured | Effect |
|---|---|---|---|
| Use FAnDa instead of Copy+Anon | Follow-up query analysis · FollowUp test | Symbol Accuracy: 27.00 ± 4.32 -> 47.80 ± 1.14 | Improved |
| Use FAnDa instead of Copy+Anon | Follow-up query analysis · FollowUp test | BLEU: 49.43 ± 1.11 -> 59.02 ± 0.54 | Improved |
| Use FAnDa instead of Copy+Anon | Follow-up query analysis · FollowUp test | Execution Accuracy: 40.77 -> 60.19 | Improved |
| Add anonymization to CopyNet | CopyNet · FollowUp test | Symbol Accuracy: 19.30 ± 0.93 -> 27.00 ± 4.32 | Improved |
| Add intent classification instead of assigning Refine to every follow-up (ablation) | FAnDa · FollowUp test | Symbol Accuracy: 35.30 ± 0.44 -> 47.80 ± 1.14 | Improved |
| Add intent classification instead of assigning Refine to every follow-up (ablation) | FAnDa · FollowUp test | BLEU: 55.01 ± 0.86 -> 59.02 ± 0.54 | Improved |
| Use the ranking model instead of randomly selecting a segment sequence (ablation) | FAnDa · FollowUp test | Symbol Accuracy: 24.30 ± 6.70 -> 47.80 ± 1.14 | Improved |
| Use the ranking model instead of randomly selecting a segment sequence (ablation) | FAnDa · FollowUp test | BLEU: 52.92 ± 2.24 -> 59.02 ± 0.54 | Improved |
| Add supervised pretraining on manually annotated tags from 100 training triples (ablation) | FAnDa · FollowUp test | Symbol Accuracy: 47.80 ± 1.14 -> 48.20 ± 1.02 | Improved |
| Add supervised pretraining on manually annotated tags from 100 training triples (ablation) | FAnDa · FollowUp test | BLEU: 59.02 ± 0.54 -> 59.87 ± 0.43 | Improved |
Closest papers by meaning
- Text2Analysis: A Benchmark of Table Question Answering with Advanced Data Analysis and Unclear Queries (AAAI 2024 · similarity 0.84)
- SQLens: An End-to-End Framework for Error Detection and Correction in Text-to-SQL (NeurIPS 2025 · similarity 0.83)
- CogSQL: A Cognitive Framework for Enhancing Large Language Models in Text-to-SQL Translation (AAAI 2025 · similarity 0.83)
- UQE: A Query Engine for Unstructured Databases (NeurIPS 2024 · similarity 0.82)
- Structure-Guided Large Language Models for Text-to-SQL Generation (ICML 2025 · similarity 0.82)
- Confidence Estimation for Error Detection in Text-to-SQL Systems (AAAI 2025 · similarity 0.82)
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.
