Papers › MAG-SQL: Multi-Agent Generative Approach with Soft Schema Linking and Iterative...

MAG-SQL: Multi-Agent Generative Approach with Soft Schema Linking and Iterative Sub-SQL Refinement for Text-to-SQL

15 Aug 2024arXiv:2408.07930archive 2025-07-28

Wenxuan Xie, Gaochen Wu, BoWen Zhou

Recent In-Context Learning based methods have achieved remarkable success in Text-to-SQL task. However, there is still a large gap between the performance of these models and human performance on datasets with complex database schema and difficult questions, such as BIRD. Besides, existing work has neglected to supervise intermediate steps when solving questions iteratively with question decomposition methods, and the schema linking methods used in these works are very rudimentary. To address these issues, we propose MAG-SQL, a multi-agent generative approach with soft schema linking and iterative Sub-SQL refinement. In our framework, an entity-based method with tables' summary is used to select the columns in database, and a novel targets-conditions decomposition method is introduced to decompose those complex questions. Additionally, we build a iterative generating module which includes a Sub-SQL Generator and Sub-SQL Refiner, introducing external oversight for each step of generation. Through a series of ablation studies, the effectiveness of each agent in our framework has been demonstrated. When evaluated on the BIRD benchmark with GPT-4, MAG-SQL achieves an execution accuracy of 61.08%, compared to the baseline accuracy of 46.35% for vanilla GPT-4 and the baseline accuracy of 57.56% for MAC-SQL. Besides, our approach makes similar progress on Spider. The codes are available at https://github.com/LancelotXWX/MAG-SQL.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2408.07930")

Code

Syntology Ran 21 of 25 code samples harvested from 1 repository linked to this paper; 4 have no recorded run. Of those that ran: 1 ran · honoured contract; 3 ran · violated contract; 2 ran · our draft was wrong; 15 ran with no contract checked.

By repository: official repository: 25 samples from 1 repository, 21 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

LancelotXWX/MAG-SQL officialmentioned in papermentioned on GitHubMIT report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

25 samples harvested; 21 ran; 1 honoured the contract we drafted; 4 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

1ran · honoured contract
3ran · violated contract
2ran · our draft was wrong
15ran
4unverified

Licence: 0 of the 25 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from LancelotXWX/MAG-SQL. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

condition_has_like LancelotXWX/MAG-SQL/evaluation/evaluation_spider.py official repository ran · violated contract MIT (permissive) · 8231e3fdffa9c8ee · report
condition_has_or LancelotXWX/MAG-SQL/evaluation/evaluation_spider.py official repository ran · violated contract fingerprinted MIT (permissive) · f2d76234528d3b57 · report
condition_has_sql LancelotXWX/MAG-SQL/evaluation/evaluation_spider.py official repository ran · violated contract MIT (permissive) · 954c76e256cc7220 · report
execute_sql LancelotXWX/MAG-SQL/evaluation/evaluation_bird_ex.py official repository ran · honoured contract MIT (permissive) · 9c222961369b8d61 · report
execute_sql LancelotXWX/MAG-SQL/evaluation/evaluation_bird_ves.py official repository ran MIT (permissive) · 9a2dafca24381ab4 · report
get_schema LancelotXWX/MAG-SQL/evaluation/process_sql.py official repository ran fingerprinted MIT (permissive) · 0992d2a47ae733fa · report
get_schema_from_json LancelotXWX/MAG-SQL/evaluation/process_sql.py official repository ran MIT (permissive) · c4dcaf29b663a364 · report
is_commonword LancelotXWX/MAG-SQL/main_scripts/bridge_content_encoder.py official repository ran MIT (permissive) · b724cc2d4066a429 · report
is_number LancelotXWX/MAG-SQL/main_scripts/bridge_content_encoder.py official repository ran MIT (permissive) · f39fdc1ce6cffe08 · report
is_stopword LancelotXWX/MAG-SQL/main_scripts/bridge_content_encoder.py official repository ran MIT (permissive) · d56a1f9221823a48 · report
is_valid_date LancelotXWX/MAG-SQL/main_scripts/utils.py official repository ran MIT (permissive) · e683610fb710931f · report
is_valid_date_column LancelotXWX/MAG-SQL/main_scripts/utils.py official repository ran fingerprinted MIT (permissive) · 6e6792ab3f04a268 · report
iterated_execute_sql LancelotXWX/MAG-SQL/evaluation/evaluation_bird_ves.py official repository ran MIT (permissive) · 2b57439bdcb8e43d · report
join_tokens LancelotXWX/MAG-SQL/evaluation/parse.py official repository ran MIT (permissive) · 2178b95c919ae420 · report
load_json LancelotXWX/MAG-SQL/evaluation/evaluation_bird_ex.py official repository ran · our draft was wrong MIT (permissive) · 72eccdcbaaff9c8d · report
permute_tuple LancelotXWX/MAG-SQL/evaluation/exec_eval.py official repository ran MIT (permissive) · f23d6ff3772963bc · report
postprocess LancelotXWX/MAG-SQL/evaluation/parse.py official repository ran fingerprinted MIT (permissive) · 58e472cff9780d63 · report
quick_rej LancelotXWX/MAG-SQL/evaluation/exec_eval.py official repository ran MIT (permissive) · a40510ba6a37f798 · report
rename_file LancelotXWX/MAG-SQL/main_scripts/utils.py official repository ran MIT (permissive) · 009030dd773225d2 · report
replace_multiple_spaces LancelotXWX/MAG-SQL/evaluation/evaluation_bird_ex.py official repository ran · our draft was wrong fingerprinted MIT (permissive) · 13917dfdddd70f5e · report
unorder_row LancelotXWX/MAG-SQL/evaluation/exec_eval.py official repository ran MIT (permissive) · afc8ef89576ef2aa · report
api_func LancelotXWX/MAG-SQL/main_scripts/llm.py official repository unverified MIT (permissive) · 61ca5e74d54e4c09 · report
clean_abnormal LancelotXWX/MAG-SQL/evaluation/evaluation_bird_ves.py official repository unverified MIT (permissive) · 3553bf8903f0b994 · report
safe_call_llm LancelotXWX/MAG-SQL/main_scripts/llm.py official repository unverified MIT (permissive) · a0e3cdde26258b5d · report
tokenize LancelotXWX/MAG-SQL/evaluation/process_sql.py official repository unverified MIT (permissive) · a906f5de1f5971b0 · report

Tasks

In-Context LearningText to SQLText-To-SQL

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Methods

Absolute Position EncodingsAdamAttentionBPEDense ConnectionsDropoutGPT-4Label SmoothingLayer NormalizationLinear LayerMulti-Head AttentionPosition-Wise Feed-Forward LayerResidual ConnectionSoftmaxTransformer

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections