Code Generation

Follow this topic Ask about this topic

Code generation asks how models can turn requests, examples, images, or formal specifications into programs that work in real settings, from short functions to whole repositories and domain-specific software. Common approaches include prompting and fine-tuning, retrieval of relevant code or documentation, constrained decoding, and agents that iteratively run, test, and repair their outputs. Multimodal systems also use rendered interfaces, charts, and diagrams as input or feedback, while reinforcement learning can use execution tests, compilers, or formal verifiers as training signals. Benchmarks such as HumanEval, MBPP, LiveCodeBench, and repository-level tasks show progress, but passing tests or compiling does not guarantee that code meets its intended behavior, visual requirements, security needs, or deployment constraints.

Latest papers

313 papers42 papers in the last four weeks, up 68% on the four weeks before. 0.2% of all new papers.

Green Code or Costly Code? Evaluating the Energy Costs of Iterative LLM-Based Code Optimizations
Oct 23, 2026Sadra Emami, Benjamin Weigell, Mohammad Reza Mousavi +3Large Language Model-Guided Optimization Code Generation
University of Augsburg, Augsburg, Germany · King’s College London, London, United Kingdom
Shorter Is Not Cheaper: A Controlled Comparison of Declarative Animation Specifications and Free-Form Code for LLM-Generated Explanations
Oct 23, 2026Jiantang HuangStructured Output Generation Code Generation
RIPPLE: Recovery Internalization through Process-Level Policy Learning and Effect-Aware Credit for Long-Horizon Agentic Coding
Oct 22, 2026Jingxuan Xu, Zihang Xu, Yuechen Luo +9Code Generation RL for LLM Agents
National University of Singapore, Singapore · Shanghai Artificial Intelligence Laboratory, Shanghai, China · Tsinghua University, Beijing, China +1
ProductWebBench: Evaluating Website Continuity Beyond Task Completion
Oct 22, 2026Jiajun Wu, Jian Yang, Yaxin Du +9Software Engineering Benchmarks Code Generation Evaluation
Beihang University · IQuest Research · Huawei Technologies Ltd.
Did the Problem Actually Get Harder? Testing LLMs on Paired P and NP-Complete Problems Hardened by Adversarial Multi-Agent Rewriting
Oct 21, 2026Nurul Labib Sayeedi, Md. Faiyaz Abdullah Sayeedi, Zabir Al Nazi +2Language Model Reasoning Evaluation Agent Benchmarks
United International University · BRAC University · University of California, Riverside +1
From Abstraction to Action: DSLs for Natural Language Abstraction Interpretation and Execution
Oct 20, 2026Amit Lisha, Reut TsarfatyInstruction Following Structured Reasoning in Language Models
Bar-Ilan University, Israel
From Code Generator to Engineering Companion: A Plant-aware Agent for PLC Programming
Oct 20, 2026Joschka Kersting, Gesa BenndorfCoding Agents Code Generation
Fraunhofer IOSB-INA Lemgo, Germany
Self-Decomposing Diffusion for Code Generation
Oct 20, 2026Lize Shao, Zichen Xie, Zhizhen Chen +2Code Generation LLM Code Generation
University of Virginia
Revision-Based Learning from Interactive Experience and Feedback
Oct 19, 2026Jiawen Li, Chenming Ge, Alice Gao +3Code Generation Web Application Generation
Purdue University · University of Michigan · University of Washington +1
VHDLBench: Benchmarking and Fine-Tuning Large Language Models for VHDL-Based Chip Design
Oct 19, 2026Patrik Drazic, Benaoumeur Senouci, Boualem BenatallahElectronic Design Automation Code Generation
University of Southern Denmark, Sønderborg, Denmark · RINN AI-Research Ireland Centre for AI and Data Science, Dublin City University Dublin, Ireland
Learn the Interface, Not the Implementation: Output Representation Dominates Training Regime in LLM-Based RTL Generation
Oct 19, 2026Siddharth Patel, Vikram Singh, Siddharth Savyasachi Malu +1RTL Code Generation Code Generation
Dept. of Electronics Engineering, Shiv Nadar Institution of Eminence, Delhi-NCR, India · STMicroelectronics, Greater Noida, India · AI Center of Excellence, Shiv Nadar Institution of Eminence, Delhi-NCR, India
Backend-Frontend Converter: Component-Based Conversion of .NET Computational Models to TypeScript
Oct 17, 2026Petr Ivankov, Max ShestovCode Translation Code Generation
Independent researcher · Georgia Institute of Technology
PrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence through Decomposed Rubric Synthesis
Oct 14, 2026Qiushi Sun, Jingyang Gong, Yang Liu +14Multimodal Large Language Models Code Generation
The University of Hong Kong · Nanjing University · Fudan University +3
Verifier-Grounded Agentic RL for Provably Correct Code
Oct 13, 2026George Ma, Yuhao Zhang, Bing He +5RL for Code Generation Verified Code Generation
UC Berkeley · AWS AI Labs
AesCode: Aesthetic Code Generation with Decoupled Cross-Modal Rewards
Oct 13, 2026Xin Gao, Yifan Yang, Xin Li +13Code Generation RL for Code Generation
Yale University · Microsoft · Nanyang Technological University +3
SolBehave: Evaluating the Behavioural Correctness of LLM-Generated Smart Contracts
Oct 13, 2026Giovanni Pio Delvecchio, Stefano Fantazzini, Giacomo Frisoni +4Code Generation Evaluation LLM Code Generation
University of Bologna, Bologna, Italy · University of Copenhagen, Copenhagen, Denmark
SpecCloak: Probing Harmful Code Generation through Engineering Specifications
Oct 13, 2026Wenbin Chen, Jiexin Wang, Guanting Wu +2LLM Security LLM Code Generation
South China University of Technology
Deleting Code Without Losing Knowledge: Regenerative Software for Python Libraries
Oct 12, 2026Thomas Bartz-BeielsteinAutomated Software Engineering Code Generation
EvoWeb: A Benchmark for Coding Agents on Evolving Web Applications
Oct 12, 2026David Nukrai, Bar Alon, Nativ Levy +6Code Generation Benchmarks Web Application Generation
Stealth Startup, Tel Aviv · Tel Aviv University
Passing Tests, Missing Semantics: Evidence of Interpretive Lock-In in Plan-Based DataFrame Code Generation
Oct 11, 2026Jingjing Liu, Dishu Yang, Hongbo Qin +1Code Generation Code Generation Evaluation
University of California, Berkeley, Berkeley, CA, USA · Northeastern University, San Jose, CA, USA · Northeastern University, Boston, MA, USA
Conditional Transfer from Controlled Pretraining Mixtures to Code
Oct 8, 2026Ohad RubinSynthetic Data Pretraining Data Mixture Optimization
TaoD2C-Bench: Benchmarking MLLMs for Industrial UI Code Generation Beyond Visual Fidelity
Oct 7, 2026Chengwei Shi, Yunnong Chen, Tingting Zhou +5Image-to-Code Generation Code Generation
Zhejiang University · Taobao & Tmall Group of Alibaba
Visible Reasoning Is Not a Universal Optimizer: Persona- and Thinking-Dependent Effects in Analytics Code Generation
Oct 7, 2026Bhawani Shankar Leelar, Pawan Chorasiya, Davin Hill +2CoT Reasoning LLM Code Generation
Optum AI
From Pixel to Coding: Evaluating the Figure Reproduction Capabilities of MLLMs
Oct 7, 2026Zijian Chen, Zhengyu Chen, Bohan Liang +7Multimodal Large Language Models Multimodal Model Evaluation
Institute of Image Communication and Information Processing, Shanghai Jiao Tong University, Shanghai, 200240, China. · Shanghai Artificial Intelligence Laboratory, Shanghai, 200030, China. · Macao Polytechnic University, Macao, 999078, China.
Adaptive Code Generation for Controlling Robots
Oct 7, 2026Justus Flerlage, Thorsten Wittkopp, Alexander Acker +1Code Generation for Robotics Language Model-Based Control
Technische Universität Berlin, Germany · logsight.ai, Berlin, Germany
SPLATIFY: Reproduce, Discover, Innovate! From Papers and Ideas to Trainable 3DGS Code
Oct 6, 2026Seemandhar Jain, Keshav Gupta, Manmohan ChandrakerCode Generation 3D Gaussian Splatting
University of California, San Diego
WorldSolver: Can LLM Agents Simulate the Physical Dynamics via Solver Generation?
Oct 6, 2026Siru Jiang, Yongzhe Lyu, Shuo Lu +6Physics-Based Simulation Code Generation
NLPR & MAIS, CASIA · PKU · Huawei Noah’s Ark Lab
SPECTRUM: Proximal Spectral Modulation for Looped Self-Distillation
Oct 5, 2026Yunbo Long, WenJie Chen, Jiaquan Zhang +4Self-Distillation Code Generation
University of Cambridge · Fudan University
VHDL-REPOBENCH: A Repository-Level Benchmark for Evaluating Large Language Models on VHDL Design Generation
Oct 4, 2026Prashanth Vijayaraghavan, Akul Malhotra, Ashutosh Jadhav +2LLM Evaluation Code Generation
IBM Research San Jose, CA, USA
Spec2Game: Can LLMs Generate Complete Playable Games from Detailed Specifications?
Oct 3, 2026Yixue Cai, Yuzhe Zhao, Hanxiang Chao +4Code Generation LLM Code Generation
The Chinese University of Hong Kong Hong Kong, China · Nankai University Tianjin, China · Wuhan University Wuhan, China

Oct 23, 2026J/K move · Enter open · S save

Green Code or Costly Code? Evaluating the Energy Costs of Iterative LLM-Based Code Optimizations

Sadra Emami, Benjamin Weigell, Mohammad Reza Mousavi, Gunel Jahangirova+2

University of Augsburg, Augsburg, Germany · King’s College London, London, United Kingdom

Large Language Model-Guided Optimization Code Generation Energy-Aware LLM Inference LLM Self-Refinement

Open paper AskPDF arXiv

Large Language Models (LLMs) are increasingly used to optimize code for performance. This practice often assumes that improvements in execution speed naturally translate into overall lower energy consumption. We challenge this assumption empirically by measuring whether performance-optimized code produced by LLM self-refinement pipelines is also energy-efficient relative to canonical human reference solutions, and whether the energy cost of the optimization process can be offset by subsequent execution-time savings. We evaluate 11 open-source models across four decoding configurations on all 1,000 problems of EffiBench, a state-of-the-art performance benchmark, while collecting per-process, hardware-level energy measurements throughout. Inference energy spans 0.79 to 8.68 MJ across model-variant pairs, which is roughly equivalent to the energy of 250 to 2800 GPT-4.1 queries. Limiting the maximum number of output tokens has the largest impact, reducing energy use by up to 69% compared to the default configuration. On the execution side, 42 of 44 configurations produce code that consumes more energy per problem than the canonical human reference, with overheads ranging from +4% to +886%. The two exceptions achieve only marginal gains, with break-even points on the order of 10^6 to 10^7 executions. Mixture-of-Experts architectures consume substantially less inference energy than equivalently-sized dense models without sacrificing correctness, making architectural choice a more consequential sustainability lever than decoding parameters. Performance optimization and energy efficiency are not always aligned goals: LLM self-refinement consistently produces faster-seeking but energy-heavier code relative to what a human engineer would write. All artifacts are available at https://doi.org/10.5281/zenodo.23598170.