Excited to share that our recent work DivSkill-SQL is now ranked #1 on the official Spider2-Lite leaderboard🎉 Text-to-SQL agent ensembling, which generates multiple SQL candidates, then selects the best one, has been shown to be effective.
15141.5K
Posts avec #Text2SQL
#Text2SQL
Les posts publics sur X qui utilisent #Text2SQL.
Hashtags associés
- @J10ngl1ZOur idea is to optimize the ensemble more directly for Pass@K. Specifically, we aim to learn residual skills that capture the missing capabilities needed to address what existing agents fail to solve. This leads to more diverse, targeted, and effective Text-to-SQL ensembles.103176
- @J10ngl1ZHuge thanks to collaborators at Snowflake AI Research and UCSD: @howardghqqq, @parjanyapp, @nikkiklj1, @BaharanKhatami, @XuCanwen, @Xiaodong_Yu_126, @YingyuL2, @yao_zhewei, @YuxiongH, Babak Salimi! Our paper: #Text2SQL #Agents #LLM #Databases103266
MyScale@MyScaleDBText-to-SQL combined with retrieval augmentation is taking #LLM-powered apps to new heights. @MyScaleDB, a SQL #vectordatabase, supports both #Text2SQL and vector queries. ❓Wanna see how we automate this process? Try out ChatData now👉003242
- MyScale@MyScaleDB🥳Introducing an excellent update on ChatData! Wikipedia is now available as a knowledge base! 😉Come unlock the true potential of information retrieval with ChatData and discover a world of knowledge at your fingertips. #selfquery #text2sql #vectorsql #vectordatabase0296K
Tanmay Parekh@tparekh97Proposing a new “Parallel Exploration” Agentic framework PExA for complex code generation, specifically text-to-SQL. We achieve one of the best numbers on the Spider 2.0 leaderboard. Paper incoming on ArXiv soon! #agenticAI #text2sql #llmTech At Bloomberg @TechAtBloombergA team of researchers from Bloomberg’s #AI Engineering group introduced PExA, an #agenticAI framework that achieved 70.2% execution accuracy on the Spider 2.0 leaderboard, one of the most demanding benchmarks for #text2sql generation #CodeGeneration (1/2)2191.2K
Tech At Bloomberg@TechAtBloombergA team of researchers from Bloomberg’s #AI Engineering group introduced PExA, an #agenticAI framework that achieved 70.2% execution accuracy on the Spider 2.0 leaderboard, one of the most demanding benchmarks for #text2sql generation #CodeGeneration (1/2)1091.8K
It's easy to generate a 2022 summary of a #GitHub repository with #OpenAI: 1. Open (or choose another repo) 2. Ask questions with natural language 3. Run SQL (generated by AI) to get result Q1: How many pull requests open in 2022 of this repo #OpenSource
34133.2K- @OSSInsightQ2: How many stars (WatchEvent) this year of this repo A: 3572 #OpenSource #OpenAI #text2sql #GitHub #AnnualReport2022
102513 
Till Döhmen@tdoehmenEver wondered how well #Claude, #ChatGPT, and other models can quack DuckDB SQL? 🦆 Check out our brand new @duckdb #Text2SQL benchmark space on @huggingface and find out!
18304.5K- Learn more about using LangChain to build a RAG text-to-SQL solution, the challenges you will face when moving to production use cases, and how you can solve them! 🚀 Great post written by @chilijung! #langchain #text2sql #sql #ai #gpt #openai #rag #pipeline Check it out:
013397 - #QueryWeaver #MCP is out! Talk with your database(s) with this open-source #Text2SQL tool 👇 Check it out: See it in action today (9.9):013278
Steve.Shi@database_llmA very nice survey of alignment research, we are considering aligning llm generated sqls with its execution cost. #llms #text2sql #SQLelvis @omarsar0If you are interested in reviewing research on alignment for LLMs, look no further. This new survey paper covers different aspects of alignment research such as data collection, training methodologies, and model evaluation. Aligning LLMs with human expectations is an active area of research but is already responsible for some of the most advanced LLMs we have today. This paper sheds light on this area of research, tactics, and insights, including challenges and future directions. It's an impressive resource for researchers and practitioners. There is also a nice GitHub repo along with the paper. paper: repo:022335
If you are interested in reviewing research on alignment for LLMs, look no further. This new survey paper covers different aspects of alignment research such as data collection, training methodologies, and model evaluation. Aligning LLMs with human expectations is an active area of research but is already responsible for some of the most advanced LLMs we have today. This paper sheds light on this area of research, tactics, and insights, including challenges and future directions. It's an impressive resource for researchers and practitioners. There is also a nice GitHub repo along with the paper. paper: repo:
6240901122K
Davood Rafiei@DavoodRafieiOur #EMNLP2023 main paper "Evaluating Cross-Domain Text-to-SQL Models and Benchmarks" is now at The evaluation includes major #text2sql benchmarks, showing some of the areas for improvement. #spider #bird107364- Davood Rafiei@DavoodRafieiOur paper on "Evaluating Cross-Domain Text-to-SQL Models and Benchmarks" is accepted to EMNLP 2023. The evaluation includes #spider and #bird. Congrats to M. Pourreza for the great work. It is coming to arXiv soon. #text2sql206700
- Afficher plus de posts
