Program Synthesis

Ask about this topic

Overview

Program synthesis studies how to turn desired behavior into executable code, from small transformations and algorithms to robot policies and engineering designs. Methods include enumerative and constraint-based search over domain-specific languages, learning programs from input-output examples, and language-model proposals refined through execution, counterexamples, or formal verification. These approaches can produce reusable, inspectable programs and support settings such as CAD, GPU optimization, and verified software; benchmarks such as ARC-AGI, PBEBench, and ProgramBench test how well systems infer behavior and generalize. Systems are increasingly effective when tasks have clear constraints and feedback, but open-ended synthesis remains difficult: plausible code may fail behavioral tests, miss the intended specification, or generalize poorly beyond the examples and search budget.

Latest papers

CardsList

181 papers29 papers in the last four weeks, up 61% on the four weeks before. 0.2% of all new papers.

  1. RoboAuto: Autonomous Robot Program Search through Multi-Level Physical Understanding

    Oct 27, 2026Buzhao Liu, Jianing Ye, Xiu Yuan +6Large Language Model-Based Robot PlanningRobotic Manipulation

    Washington University in St. Louis

  2. Executable Archaeology of Programs That Model Programming: Reanimating the Heuristic Compiler (1961) and KBEmacs (1984)

    Oct 26, 2026Jeff Shrager, Kent M. Pitman, Mia GolecComputational Cognitive ModelingProgram Synthesis

  3. Ranking Function Generation with LLMs for Termination

    Oct 26, 2026Rongkang Li, Jiaxin Hu, Taolue Chen +3Program SynthesisLLM-Based Program Synthesis

    Key Laboratory of System Software (Chinese Academy of Sciences), China · Institute of Software, Chinese Academy of Sciences, China · University of Chinese Academy of Sciences, China +2

  4. FlashScript: Compiling Recurring Agent Workflows into Reusable Neuro-Symbolic Programs

    Oct 24, 2026Sadra Sabouri, Hosein Hasanbeig, Adam Sage +6LLM Agent Workflow OptimizationLLM-Based Program Synthesis

    University of Southern California, Los Angeles, CA · Microsoft, Redmond, WA

  5. AutoFraming: Synthesizing Dynamic Frames of C Code with LLMs

    Oct 21, 2026Han Zhou, Yu Luo, Dianxiang XuProgram SynthesisFormal Verification

    University of Missouri–Kansas City, Kansas City, USA · University of Central Missouri, Warrensburg, USA

  6. Policy Search in Code Space for Multi-Agent Pathfinding

    Oct 19, 2026Matthias Eder, Pantelis Emmanouil, Vivek Farias +7Multi-Agent Path FindingProgram Synthesis

    KNAPP · MIT · Columbia University +1

  7. Persistent Memory and Iterative Improvement in an LLM Agent for Chinese-Character Drawing

    Oct 16, 2026Peilin Wu, Yi-Zhuang YouPersistent Memory for Language ModelsContinual Learning for LLM Agents

    Palo Alto High School · University of California San Diego

  8. Scaling Online Exploration for Robot Agents

    Oct 15, 2026Runhan Huang, Xiangcheng Zhang, Haoyu Chen +17Robot LearningContinual Robot Learning

    Harvard University · NVIDIA · UC Berkeley +2

  9. RenoLayout: Unified Floorplan Renovation and Furniture Layout for Constraint-Aware Residential Design

    Oct 15, 2026Jingguo Liu, Yuanqing Liu, Chong Zhang +1Constrained Generative ModelingIndoor Scene Generation

    Beike

  10. Instruction-Conditioned Motion Control for Robotic Arms: Language-to-Code Decomposition and Reinforcement Learning

    Oct 14, 2026Fan Huang, Dezhong Tong, Changyu Chen +2Language-Conditioned Robot ManipulationLanguage-Conditioned Robot Control

    Indiana University Bloomington · University of Michigan · Singapore Management University

  11. Write Once, Plan Many: Building Reusable Visual Planning Systems from One Scene

    Oct 13, 2026Ruixiao Yang, Yilun Hao, Mingxin Yu +2Program SynthesisModel-Based Planning

    MIT · MIT-IBM Watson AI Lab

  12. Compiling Expert Procedures into Verified Programs: When a Small Model Is Enough

    Oct 13, 2026Vishvesh Bhat, Asad Aali, Muhammad Ahmed MohsinLLM Tool UseLLM-Based Program Synthesis

    CoreThink AI · Stanford University

  13. Fly-by-Code: Embodied Coding Agents for Aerial Manipulation with Active Visual and Physical Feedback

    Oct 8, 2026Jaewoo Lee, Jeongyeon Seo, Sihyun Cho +10Error FeedbackAerial Manipulation

    Seoul National University · Carnegie Mellon University · University of Maryland, College Park +2

  14. Long Text to Predictive Features: LLM-Guided Blockwise Feature Engineering via Executable Program Search

    Oct 8, 2026Ziming Dai, Dabiao Ma, Ziheng Guo +2Automated Feature EngineeringLarge Language Model-Based Feature Engineering

    City University of Hong Kong Hong Kong, China

  15. EliteTable: Self-Improving Program Search for Executable Control Policies

    Oct 8, 2026Stuart FarmerProgram SynthesisLLM-Based Program Synthesis

    Casa Dilauro Independent Research (casadilauro.com)

  16. Learning How to Search for Plans with Exponentially Less Space

    Oct 7, 2026Dominik Drexler, Simon Ståhlberg, Markus Fritzsche +1Symbolic PlanningPolicy Learning

    Linköping University Linköping, Sweden · RWTH Aachen University Aachen, Germany · Universidad Simón Bolívar Caracas, Venezuela

  17. EvoSignal: LLM-Guided Evolutionary Design of Modular Traffic Signal Control Programs

    Oct 7, 2026Leizhen Wang, Peibo Duan, Zhenlin Qin +5Traffic Signal ControlAutomated Heuristic Design

    Monash University · KTH Royal Institute of Technology · Chinese Academy of Sciences +2

  18. CADFather: Autonomous CAD Reconstruction through Coordinated Tool Use

    Oct 6, 2026Gennadiy Savrasov, Maksim Elistratov, Nikita Gavrilov +7Parametric CAD Modeling3D Reconstruction

    Lomonosov Moscow State University · Innopolis University · FusionBrain Lab

  19. Learning Explainable Representations of Complex Game-playing Strategies

    Oct 6, 2026Abhijeet Krishnan, Colin M. Potts, Arnav Jhala +3Game-Playing AgentsProgram Synthesis

    Venture IV 420, North Carolina State University, Raleigh, NC 27606 USA · Department of Aerospace Engineering, Indian Institute of Technology Bombay, Powai, Mumbai 400076, Maharashtra, India · Tata Research Development and Design Centre, Pune 411013, Maharashtra, India +1

  20. CheckerBench: Can Long-Horizon Agents Synthesize Static-Analysis Checkers?

    Oct 6, 2026Hang He, Li Wang, Hao Chen +9Software Engineering AgentsStatic Code Analysis

    East China Normal University, Humanlaya Data, Shanghai Jiao Tong University, Peking University, Shanghai Innovation Institute

  21. Teaching a Minimalist Machine to Discover Recursive Programs for Arithmetic

    Oct 5, 2026Dominik Magiera, Christiane Wiebel-Herboth, Frank JäkelCognitive ModelingProgram Synthesis

    Centre for Cognitive Science & Institute of Psychology, Technische Universität Darmstadt, Germany · Honda Research Institute Europe GmbH, Offenbach am Main, Germany

  22. CADForge: Agentic Single-View CAD Reconstruction with Explicit Geometry Reasoning

    Oct 3, 2026Keyang Lu, Zhifei Yang, Tianao Dong +3Parametric CAD ModelingSingle-View 3D Reconstruction

    Peking University · Shenzhen Loop Area Institute · Beijing Normal University

  23. 4DCodeBench: Benchmarking Agents on Inverse Graphics of Dynamic Scenes

    Oct 2, 2026Ruihong Shen, Žiga Kovačič, Peter Kulits +6Dynamic Scene ReconstructionAI Agent Benchmarks

    Johns Hopkins University · Stanford University · Max Planck Institute for Intelligent Systems +1

  24. WAMpy: Efficient Synthesis of Prolog Programs in Python

    Oct 2, 2026Dominik Magiera, Lukas Röhrig, Frank JäkelProgram SynthesisLogic Programming

    Centre for Cognitive Science & Institute of Psychology, Technische Universität Darmstadt, Germany

  25. When to Compile a Computer-Use Agent? Measuring Payback and Making Compilation Decisions for Token Efficiency

    Oct 2, 2026Yulong Ming, Jie Xu, Zihan Wu +1Computer-Use AgentsLLM Agent Workflow Optimization

    City University of Hong Kong

  26. Constraint-Aware Training

    Oct 2, 2026Jinwoo KimConstrained Generative ModelingEfficient Language Model Training

    Department of Computer Science and Engineering University of California-San Diego La Jolla, CA

  27. Law And Order: Tax Law Autoformalization

    Oct 2, 2026Sophia Simeng Han, Yoshiki Takashima, Anjiang Wei +2AutoformalizationProgram Synthesis

    Stanford University · Stanford Law School · Pramaana Labs +1

  28. StepCAD: Mesh-to-CAD Code Generation via LLM Policy and Geometry-Guided Search

    Oct 1, 2026Ghadi Nehme, Faez AhmedParametric CAD ModelingProgram Synthesis

    Department of Mechanical Engineering Massachusetts Institute of Technology Cambridge, MA 02139

  29. InfoAgent: Traceable Generation and Repair of Evidence-Grounded Infographics

    Sep 30, 2026Yifan Li, Tong Li, Qi Zeng +6Data VisualizationData Provenance

    University of Science and Technology of China · Meituan · Peking University, Peking University

  30. MemEvo: Automatic Discovery of Streaming Video Memory Mechanisms

    Sep 29, 2026Guohong Liu, Jialei Ye, Shanhui Zhao +2Memory-Augmented VLMsLanguage Model-Guided Search Heuristics

    Institute for AI Industry Research, Tsinghua University · Peking University