Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Soni, Chinmay, Chourasia, Shivam, Kumar, Gaurav, Kapoor, Hitesh
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2603.24023
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915889452941312
author Soni, Chinmay
Chourasia, Shivam
Kumar, Gaurav
Kapoor, Hitesh
author_facet Soni, Chinmay
Chourasia, Shivam
Kumar, Gaurav
Kapoor, Hitesh
contents Applying large, proprietary API-based language models to text-to-SQL tasks poses a significant industry challenge: reliance on massive, schema-heavy prompts results in prohibitive per-token API costs and high latency, hindering scalable production deployment. We present a specialized, self-hosted 8B-parameter model designed for a conversational bot in CriQ, a sister app to Dream11, India's largest fantasy sports platform with over 250 million users, that answers user queries about cricket statistics. Our novel two-phase supervised fine-tuning approach enables the model to internalize the entire database schema, eliminating the need for long-context prompts. This reduces input tokens by over 99%, from a 17k-token baseline to fewer than 100, and replaces costly external API calls with efficient local inference. The resulting system achieves 98.4% execution success and 92.5% semantic accuracy, substantially outperforming a prompt-engineered baseline using Google's Gemini Flash 2.0 (95.6% execution, 89.4% semantic accuracy). These results demonstrate a practical path toward high-precision, low-latency text-to-SQL applications using domain-specialized, self-hosted language models in large-scale production environments.
format Preprint
id arxiv_https___arxiv_org_abs_2603_24023
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Schema on the Inside: A Two-Phase Fine-Tuning Method for High-Efficiency Text-to-SQL at Scale
Soni, Chinmay
Chourasia, Shivam
Kumar, Gaurav
Kapoor, Hitesh
Computation and Language
Artificial Intelligence
Applying large, proprietary API-based language models to text-to-SQL tasks poses a significant industry challenge: reliance on massive, schema-heavy prompts results in prohibitive per-token API costs and high latency, hindering scalable production deployment. We present a specialized, self-hosted 8B-parameter model designed for a conversational bot in CriQ, a sister app to Dream11, India's largest fantasy sports platform with over 250 million users, that answers user queries about cricket statistics. Our novel two-phase supervised fine-tuning approach enables the model to internalize the entire database schema, eliminating the need for long-context prompts. This reduces input tokens by over 99%, from a 17k-token baseline to fewer than 100, and replaces costly external API calls with efficient local inference. The resulting system achieves 98.4% execution success and 92.5% semantic accuracy, substantially outperforming a prompt-engineered baseline using Google's Gemini Flash 2.0 (95.6% execution, 89.4% semantic accuracy). These results demonstrate a practical path toward high-precision, low-latency text-to-SQL applications using domain-specialized, self-hosted language models in large-scale production environments.
title Schema on the Inside: A Two-Phase Fine-Tuning Method for High-Efficiency Text-to-SQL at Scale
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2603.24023