AI Paper Atlas

Research map / Large language models / Agents, code and math

LearNAT: Learning NL2SQL with AST-guided Task Decomposition for Large Language Models

Weibin Liao, Xin Gao, Tianyu Jia, Rihong Qiu, Yifan Zhu, Yang Lin, Xinyu Ma, Junfeng Zhao, Yasha Wang

ICLR 2026 · Large language models › Agents, code and math · approach: Mathematical and structural reasoning benchmarks · 17 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
provide manually written subtask decompositions to the LLM (ablation)Qwen2.5-Coder-32B · 500 BIRD-train examplesperformance: 30.4% upImproved
let the LLM decompose queries by prompting (ablation)Qwen2.5-Coder-32B · 500 BIRD-train examplesperformance: 3.4% upNo clear effect
use language models to verify subtask correctness by semantic similarity (ablation)Sentence-Transformer and GLM-4 · 500 subtasksverification accuracy (%): 46.8 and 36.0 total accuracyHurt
add AST-guided search and pruning to MCTS decomposition (ablation)Decomposition Synthesis Procedure · BIRD-train decompositiondecomposition success rate and token cost: 80.00% success; +8.45% over naive MCTS; 56.38% fewer tokensImproved
add self-improving adaptive demonstrations over rounds (ablation)Decomposition Synthesis Procedure · BIRD-train decompositiondecomposition success rate: +1.32% first round, then +0.4% and +0.27%Mixed
add LearNAT training (decomposition data, SFT and margin DPO) (ablation)Qwen2.5-Coder-7B · BIRD-dev totalexecution accuracy (%): 47.5 -> 58.1Improved
add AST guidance to decomposition synthesis (ablation)LearNAT · BIRD-dev total (7B)execution accuracy (%): 53.1 -> 58.1Improved
add SFT warm-up before DPO (ablation)LearNAT · BIRD-dev total (7B)execution accuracy (%): 54.5 -> 58.1Improved
add margin-aware DPO stage (ablation)LearNAT · BIRD-dev total (7B)execution accuracy (%): 53.7 -> 58.1Improved
use AST margin in DPO instead of standard DPO (ablation)LearNAT · BIRD-dev total (7B)execution accuracy (%): 56.6 -> 58.1Improved
use margin-aware DPO instead of KTO or IPO (ablation)LearNAT · BIRD-dev total (7B)execution accuracy (%): KTO 54.6, IPO 53.7 -> 58.1Improved
use task decomposition instead of CoT reasoning (ablation)LearNAT · BIRD-dev total (7B)execution accuracy (%): 49.3 -> 58.1Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy