SOTAVerified

Program Synthesis

Program synthesis is the process of automatically generating a program or code snippet that satisfies a given specification or set of requirements. This can include generating code from a formal specification, a natural language description, or example inputs and outputs. The primary goal of program synthesis is to minimize human intervention in the coding process, reduce errors, and improve productivity.

Program synthesis often involves the use of advanced algorithms, artificial intelligence, and machine learning techniques to search the space of possible programs that meet the given constraints. This process can be guided by a variety of techniques, such as constraint solving, symbolic execution, and genetic algorithms.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–50 of 423 papers

TitleStatusHype
CoRE: Enhancing Metacognition with Label-free Self-evaluation in LRMs—0
Structured Program Synthesis using LLMs: Results and Insights from the IPARC Challenge—0
Matching Markets Meet LLMs: Algorithmic Reasoning with Ranked Preferences—0
Chemical classification program synthesis using generative artificial intelligence—0
Transductively Informed Inductive Program SynthesisCode0
CLEVER: A Curated Benchmark for Formally Verified Code GenerationCode1
PoE-World: Compositional World Modeling with Products of Programmatic ExpertsCode1
Enhancing Code Generation via Bidirectional Comment-Level Mutual GroundingCode0
Rewriting Pre-Training Data Boosts LLM Performance in Math and CodeCode1
QiMeng-Xpiler: Transcompiling Tensor Programs for Deep Learning Systems with a Neural-Symbolic Approach—0
OSVBench: Benchmarking LLMs on Specification Generation Tasks for Operating System VerificationCode1
GPU accelerated program synthesis: Enumerate semantics, not syntax!—0
TinyverseGP: Towards a Modular Cross-domain Benchmarking Framework for Genetic ProgrammingCode1
DeQompile: quantum circuit decompilation using genetic programming for explainable quantum architecture search—0
CodeARC: Benchmarking Reasoning Capabilities of LLM Agents for Inductive Program Synthesis—0
Synthesizing world models for bilevel planning—0
TikZero: Zero-Shot Text-Guided Graphics Program SynthesisCode5
Research Vision: Multi-Agent Path Planning for Cops And Robbers Via Reactive Synthesis—0
Shedding Light in Task Decomposition in Program Synthesis: The Driving Force of the Synthesizer Model—0
Fully Autonomous Programming using Iterative Multi-Agent Debugging with Large Language Models—0
An Empirical Comparison of Cost Functions in Inductive Logic Programming—0
Machine Learning meets Algebraic Combinatorics: A Suite of Datasets Capturing Research-level Conjecturing Ability in Pure Mathematics—0
AutoIOT: LLM-Driven Automated Natural Language Programming for AIoT ApplicationsCode1
Factorio Learning EnvironmentCode4
GPIoT: Tailoring Small Language Models for IoT Program Synthesis and DevelopmentCode1
Program Synthesis Dialog Agents for Interactive Decision-MakingCode0
Visual Agentic AI for Spatial Reasoning with a Dynamic API—0
CODESIM: Multi-Agent Code Generation and Problem Solving through Simulation-Driven Planning and DebuggingCode2
Proving the Coding Interview: A Benchmark for Formally Verified Code Generation—0
Learning Semantics-aware Search Operators for Genetic Programming—0
QualityFlow: An Agentic Workflow for Program Synthesis Controlled by LLM Quality Checks—0
AlgoPilot: Fully Autonomous Program Synthesis Without Human-Written Programs—0
Online Prompt Selection for Program Synthesis—0
MMFactory: A Universal Solution Search Engine for Vision-Language Tasks—0
EcoSearch: A Constant-Delay Best-First Search Algorithm for Program Synthesis—0
Design2GarmentCode: Turning Design Concepts to Tangible Garments Through Program Synthesis—0
ConceptSearch: Towards Efficient Program Search Using LLMs for Abstraction and Reasoning Corpus (ARC)Code0
ARC Prize 2024: Technical ReportCode3
From Code to Play: Benchmarking Program Search for Games Using Large Language Models—0
Searching Latent Program SpacesCode2
LLMPhy: Complex Physical Reasoning Using Large Language Models and World Models—0
The Surprising Effectiveness of Test-Time Training for Few-Shot LearningCode3
Combining Induction and Transduction for Abstract ReasoningCode2
Reinforcement learning with learned gadgets to tackle hard quantum problems on real hardwareCode0
System 2 Reasoning via Generality and Adaptation—0
Mitigating Gender Bias in Code Large Language Models via Model Editing—0
Tackling the Abstraction and Reasoning Corpus with Vision Transformers: the Importance of 2D Representation, Positions, and ObjectsCode1
Can LLMs plan paths with extra hints from solvers?—0
Learning to Solve Abstract Reasoning Problems with Neurosymbolic Program Synthesis and Task Generation—0
MA-RLHF: Reinforcement Learning from Human Feedback with Macro ActionsCode1
Show:102550
← PrevPage 1 of 9Next →
All datasetsSPoC TestPSPoC TestWAlgoLisp

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1DrRepairSuccess rate @budget 10038.5—Unverified
2Multiclass localizerSuccess rate @budget 10034.2—Unverified
#ModelMetricClaimedVerifiedStatus
1DrRepairSuccess rate @budget 10057—Unverified
2Multiclass localizerSuccess rate @budget 10053.7—Unverified
#ModelMetricClaimedVerifiedStatus
1CodeTrans-MT-TF-SmallAccuracy90.31—Unverified