LLM-Based Program Synthesis

Follow this topic Ask about this topic

Momentum

22 papers in the last four weeks, up 100% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 223

All topics

Beyond Inference-Time Search: Reinforcement Learning Synthesizes Reusable Solvers
May 18, 2026Soheyl Massoudi, Gabriel Apaza, Milad Habibi +1RL for Code Generation LLM-Based Program Synthesis
ETH Zürich Zürich, Switzerland · University of Maryland College Park, MD, USA
Memory-Guided Tree Search with Cross-Branch Knowledge Transfer for LLM Solver Synthesis
May 17, 2026Fatemeh Haji, Javier Delarosa Quiros, Peyman NajafiradTree Search Large Language Model-Guided Optimization
Secure AI and Autonomy Lab · The University of Texas at San Antonio
Latent Heuristic Search: Continuous Optimization for Automated Algorithm Design
May 16, 2026Cheikh Ahmed, Mahdi Mostajabdaveh, Zirui ZhouSurrogate-Assisted Optimization Automated Algorithm Discovery
Huawei Technologies Canada, Burnaby, Canada
Property-Guided LLM Program Synthesis for Planning
May 15, 2026André G. Pereira, Augusto B. Corrêa, Jendrik SeippSymbolic Planning Counterexample-Guided Refinement
Federal University of Rio Grande do Sul · Brazil · University of Oxford +3
From I/O to Code with Discovery Agent
May 14, 2026Yihong Dong, Jiaru Qian, Haoran Zhang +7Execution-Guided Code Generation Genetic Programming
School of Computer Science, Peking University · Tongyi Lab, Alibaba Group · Wuhan University +3
SMCEvolve: Principled Scientific Discovery via Sequential Monte Carlo Evolution
May 14, 2026Jiachen Jiang, Huminhao Zhu, Zhihui ZhuLLM-Based Program Synthesis Large Language Model-Guided Scientific Discovery
Department of Computer Science and Engineering The Ohio State University
Articraft: An Agentic System for Scalable Articulated 3D Asset Generation
May 14, 2026Matt Zhou, Ruining Li, Xiaoyang Lyu +63D Asset Generation LLM-Based Program Synthesis
University of Cambridge · University of Oxford · Nanyang Technological University
Viverra: Text-to-Code with Guarantees
May 14, 2026Haoze Wu, Rocky Klopfenstein, Keith Farkas +1Code Generation LLM-Based Program Synthesis
Amherst College · Broadcom
Learning to Build the Environment: Self-Evolving Reasoning RL via Verifiable Environment Synthesis
May 14, 2026Yucheng Shi, Zhenwen Liang, Kishan Panaganti +3Reinforcement Learning RL for Language Model Reasoning
Tencent HY LLM
Distribution-Aware Programming: Learning Specialized Solvers from Experience
May 13, 2026Saharsh Koganti, Priyadarsi Mishra, Pierfrancesco Beneventano +1Large Language Model-Guided Optimization LLM-Based Program Synthesis
Texas A&M University · Massachusetts Institute of Technology
HLS-Seek: QoR-Aware Code Generation for High-Level Synthesis via Proxy Comparative Reward Reinforcement Learning
May 13, 2026Qingyun Zou, Feng Yu, Hongshi Tan +3RL for Code Generation Reward Modeling
National University of Singapore
GeoBuildBench: A Benchmark for Interactive and Executable Geometry Construction from Natural Language
May 13, 2026Jinwoong Kim, Rui Yang, Huishuai ZhangMathematical Reasoning Benchmarks LLM Grounding
Peking University · Wangxuan Institute of Computer Technology
Formalize, Don't Optimize: The Heuristic Trap in LLM-Generated Combinatorial Solvers
May 12, 2026Haoyu Wang, Yuliang Song, Tao Li +5Large Language Model-Guided Optimization LLM-Based Program Synthesis
University of Pennsylvania · University of Toronto · Google DeepMind +1
ProfiliTable: Profiling-Driven Tabular Data Processing via Agentic Workflows
May 12, 2026Wei Liu, Yang Gu, Xi Yan +5Multi-Agent LLM Systems Agentic Workflows
School of Software & Microelectronics, Peking University Beijing, China · Center for Machine Learning Research, Peking University Beijing, China Zhongguancun Academy Beijing, China
Teacher-Aware Evolution of Heuristic Programs from Learned Optimization Policies
May 11, 2026Minyu Chen, Song Qin, Ling-I Wu +2Automated Heuristic Design Evolutionary Optimization
Shenzhen Technology University · Shanghai Polytechnic University · Shanghai Jiao Tong University
Budget-Efficient Automatic Algorithm Design via Code Graph
May 11, 2026Maxime Bouscary, Manxi Wu, Saurabh AminAutomated Algorithm Discovery Large Language Model-Guided Optimization
Operations Research Center Massachusetts Institute of Technology · Department of Civil and Environmental Engineering University of California, Berkeley · Laboratory for Information & Decision Systems Massachusetts Institute of Technology
LLM4Branch: Large Language Model for Discovering Efficient Branching Policies of Integer Programs
May 11, 2026Zhinan Hou, Xingchen Li, Yankai Zhang +2Branch and Bound Mixed-Integer Linear Programming
Department of Automation, BNRist, Tsinghua University, Beijing, China.
Agentic MIP Research: Accelerated Constraint Handler Generation
May 9, 2026Liding Xu, Yugeng Zhou, Sebastian PokuttaMixed-Integer Linear Programming LLM-Based Program Synthesis
Zuse Institute Berlin, Berlin, Germany. · Independent Researcher. · Zuse Institute Berlin +1
CktFormalizer: Autoformalization of Natural Language into Circuit Representations
May 8, 2026Jing Xiong, Qi Han, Chenchen Ding +5Formal Hardware Verification Electronic Design Automation
The University of Hong Kong
Kernel Foundry: A Diagnosis-driven Evolutionary Kernel Optimizer with Multi-Experts
May 8, 2026Zixuan Huang, Da Chen, Kecheng Huang +5GPU Kernel Optimization GPU Acceleration
1The Chinese University of Hong Kong · 2Noah’s Ark Lab, Huawei
Knowledge Graphs, the Missing Link in Agentic AI-based Formal Verification
May 7, 2026Vaisakh Naduvodi Viswambharan, Keerthan Kopparam Radhakrishna, Deepak Narayan Gadde +1KG Construction Formal Hardware Verification
Infineon Technologies Dresden AG & Co. KG, Germany · Infineon Technologies Semiconductor India Private Limited, India
ReaComp: Compiling LLM Reasoning into Symbolic Solvers for Efficient Program Synthesis
May 6, 2026Atharva Naik, Yash Mathur, Prakam +2Symbolic Reasoning LLM-Based Program Synthesis
Carnegie Mellon University · Independent Researcher
CuBridge: An LLM-Based Framework for Understanding and Reconstructing High-Performance Attention Kernels
May 6, 2026Xing Ma, Yangjie Zhou, Wu Sun +6GPU Kernel Optimization GPU Acceleration
Shanghai Jiao Tong University · National University of Singapore · Shanghai Qi Zhi Institute
KernelBenchX: A Comprehensive Benchmark for Evaluating LLM-Generated GPU Kernels
May 6, 2026Han Wang, Jintao Zhang, Kai Jiang +3GPU Kernel Optimization GPU Acceleration
∗Equal contribution.
ProgramBench: Can Language Models Rebuild Programs From Scratch?
May 5, 2026John Yang, Kilian Lieret, Jeffrey Ma +9Software Engineering Agents Software Engineering Benchmarks
1Meta FAIR · 4Harvard University · 2Meta TBD +1
Strategy-Aware Optimization Modeling with Reasoning LLMs
May 4, 2026Ruiqing Zhao, Fengzhi Li, Yuan Zuo +5RL for Code Generation LLM-Based Program Synthesis
School of Computer Science and Engineering, Beihang University, Beijing, China · JIUTIAN Research, Beijing, China · MIIT Key Laboratory of Data and Decision Intelligence, Beihang University, Beijing, China
CP-SynC: Multi-Agent Zero-Shot Constraint Modeling in MiniZinc with Synthesized Checkers
May 3, 2026Yuliang Song, Eldan CohenMulti-Agent LLM Systems LLM-Based Program Synthesis
Department of Mechanical and Industrial Engineering, University of Toronto, Toronto, Canada
GA-VisAgent: A Multi-Agent application for code generation and visualization in interactive learning
May 2, 2026Wang Jian, Zhou Jianbo, Xiong Yuhao +4Code Generation LLM-Based Program Synthesis
School of Geography, Nanjing Normal University, Nanjing 210023, Jiangsu, China · School of Environment, Nanjing Normal University, Nanjing 210023, Jiangsu, China · Key Laboratory of Virtual Geographic Environment, Ministry of Education,2026 Nanjing Normal University, Nanjing 210023, Jiangsu, China +2
Towards Neuro-symbolic Causal Rule Synthesis, Verification, and Evaluation Grounded in Legal and Safety Principles
Apr 30, 2026Zainab Rehan, Christian Medeiros Adriano, Sona Ghahremani +1LLM-Based Program Synthesis Formal Verification
Hasso Plattner Institute University of Potsdam Potsdam, Germany
LLMs as ASP Programmers: Self-Correction Enables Task-Agnostic Nonmonotonic Reasoning
Apr 30, 2026Adam Ishay, Joohyung LeeLLM Self-Correction Logical Reasoning
Arizona State University · Samsung Research

May 18, 2026, cs.LGJ/K move · Enter open · S save

Beyond Inference-Time Search: Reinforcement Learning Synthesizes Reusable Solvers

Soheyl Massoudi, Gabriel Apaza, Milad Habibi, Mark Fuge

ETH Zürich Zürich, Switzerland · University of Maryland College Park, MD, USA

RL for Code Generation LLM-Based Program Synthesis Combinatorial Optimization Constrained RL RL for Combinatorial Optimization RL Fine-Tuning

Open paper AskPDF arXiv

Large language models (LLMs) typically approach combinatorial optimization as an inference-time procedure, solving each instance separately through sampling, search, or repeated prompting. We ask whether reinforcement learning can instead shift part of this reasoning cost into the weights of a code LLM, so that the model synthesizes a reusable solver for an entire problem family. We study this question on Synergistic Dependency Selection (SDS), a controlled variant of constrained Quadratic Knapsack designed to expose a specific failure mode: local signals and strict feasibility constraints make greedy heuristics attractive but unreliable. Under identical scaffolding, Best-of-64 base-model sampling saturates at an approximately 28.7% gap to the global Virtual Best Solver (VBS); code audits show that the base model often retrieves Simulated Annealing templates but misimplements the Metropolis acceptance rule. We fine-tune Qwen2.5-Coder-14B-Instruct with Group Relative Policy Optimization (GRPO) using a feasibility-gated reward and light structural scaffolding. The resulting policy converges to a constraint-aware Simulated Annealing template in 99.8% of feasible SDS outputs, achieves a 5.0% gap to that VBS, and is 91 times cheaper in post-generation execution/search cost than cumulative Best-of-64 evaluation. A compile-once check shows that one best frozen solver per seed remains highly competitive when reused unchanged across the SDS test set, while an additional-domain evaluation on Job Shop Scheduling provides narrower but positive evidence that the scaffold transfers beyond SDS. Negative ablations reveal the limits of this recipe: standard stabilizers degrade performance, a soft feasibility gate fails, and results remain sensitive to reward normalization and domain-specific design choices.

Momentum

Related topics

Latest papers 223