Генерируем SQL–запросы на локальных моделях На связи команда внутренних инициативных сервисов ecom.tech . Мы ...
#text2sql #nlp #генерация #кода #llm #sql
Origin | Interest | Match
Выжимаем максимум из опенсорсных моделей и готовим Text2SQL В любой крупной компании есть повторяющийся сцена...
#text2sql #large #language #models #open #source #llm #qwen #gpt #аналитика #данных
Origin | Interest | Match
🏆 A new milestone for our data products family.
We ranked #1 in the Spider 2.0 DBT Text2SQL benchmark.
Structured context + a semantic layer = more reliable agentic workflows.
Now opening alpha access:
Nemory → nemory.app
Databao → databao.app
#Text2SQL
Survey of Natural Language Interfaces for Databases Highlights AI‑Driven Advances
A survey of NLIDB systems describes a three‑stage pipeline and cites Spider and WikiSQL benchmarks. LLMs enable user‑friendly end‑to‑end Text2SQL and voice queries. getnews.me/survey-of-natural-langua... #nlidb #text2sql
New Bilingual Text-to-SQL Dataset Targets Process Mining Queries
The text-2-SQL-4-PM benchmark, announced Aug 2025, provides 1,655 bilingual Portuguese‑English utterances matched with 205 SQL statements for process‑mining. Read more: getnews.me/new-bilingual-text-to-sq... #text2sql #processmining
🚀 Back to #11 on 🦜 BIRD Text-to-SQL benchmark! Climbing above IBM, ByteDance & Stanford. Not done yet 💪
📜 Paper: arxiv.org/abs/2505.14174
🖥️ Code: github.com/genaasia/N-rep
Check out our #text2sql powered HeyGena app on #shopify!
Work led by @spapicchio.bsky.social , in collaboration with Simone Rossi (EURECOM) and Luca Cagliero (Politecnico Torino)
#Text2SQL #LLM #AI #NLP #ReinforcementLearning
What is your take on DuckDB vs SQLite? I started with SQLite—simple, works fine. Then explored text2sql DuckDB based on this article: bit.ly/4gPLRAc . marimo notebook: bit.ly/49U7WeY . DuckDB seems to have advantages wrt handling timestamps & text2sql? (thx)
#SQLite #DuckDB #Python #Text2SQL