English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

FERN: Cross-Benchmark Table Retrieval for Text-to-SQL with Frozen Retrievers and No Generative LLM

Forum topic · 小凯 · 2026-11-12

Summary

FERN (Feature-Engineered Ranking with No generative LLM) is a table retrieval method for Text-to-SQL systems operating over large pooled catalogs. Instead of using generative LLMs like RASL (which relies on Claude 3.5 for keyword extraction and table selection), FERN fuses three frozen retrievers into 100 candidates, describes each with up to 120 features, and reorders them with LightGBM rankers fitted once on BIRD's training questions. On Spider, where it was not fitted, FERN's candidate pool alone matches RASL's published recall at fifteen tables (98.0), and the ranker raises complete@5 from 83.9% to 95.6%. At five tables, FERN's recall is within one point of RASL on Spider (96.7 vs 97.0) and 96.6 vs 97.5 on BIRD with evidence hints. On 121 real BEAVER questions without fitting, FERN achieves recall of 82.1 at fifteen tables. On BIRD dev with evidence hints, non-retrieval features improve complete@5 by 7.7 points over retrieval-only signals. Median latency is 208 ms per question on a single A100 GPU, with no hosted model calls. Paper: arXiv:2611.14477.

Paper Overview

Field: NLP Authors: Arjun Rao, Nishant Gopinath Published: 2026-11-10 arXiv: 2611.14477

Abstract

Text-to-SQL systems over large pooled catalogs must first retrieve the tables a question needs. FERN (Feature-Engineered Ranking with No generative LLM) compares against the published table recall of RASL, which uses Claude 3.5 models to extract keywords and select tables, on pooled BIRD, Spider, and Fiben catalogs.

Method

FERN instead:

  • Fuses three frozen retrievers into 100 candidates
  • Describes each candidate with up to 120 features
  • Reorders them with LightGBM rankers fitted once on BIRD's training questions
  • Key Results

  • Spider (not fitted): The candidate pool alone reaches RASL's published recall at fifteen tables (98.0); the ranker raises complete@5 (share of questions with every gold table in the top five) from 83.9% to 95.6%
  • At five tables: FERN's recall is within one point of RASL's on Spider (96.7 vs 97.0), 96.6 vs 97.5 on BIRD with evidence hints, and 8.2 points lower on BIRD without hints (89.3)
  • BEAVER (121 real questions, not fitted): Recall at fifteen tables (82.1) falls between the two BEAVER-trained retrievers (80.8, 84.5), with complete@15 below both (49.6 vs 54.5 and 62.0)
  • Fiben: A clear gap at fifteen tables (54.2 vs 69.2, on different question sets); recall at fifty tables is close to RASL's question-only retrieval (80.7 vs 79.4)
  • Feature ablation: On BIRD dev with evidence hints, non-retrieval features improve complete@5 by 7.7 points over retrieval-only signals

Efficiency

On BIRD dev, FERN runs at a median latency of 208 ms per question on a single A100 GPU, with no calls to hosted models.

--- *Auto-collected on 2026-11-12*

Tags

#text-to-sql#table-retrieval#lightgbm#nlp#arxiv#ranking#benchmark

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178637077