Software & Coding

47 verified cases, highest-rated first.

Industries: software engineering 6, software testing 4, computer vision 2, software security 2, database systems 2, speech recognition 2, compiler software 2, software infrastructure 1, mobile security 1, cybersecurity 1, machine learning / AutoML 1, microscopy/bioimaging 1

• Faster sorting algorithms discovered using deep reinforcement learning

  Improving the efficiency of fundamental sorting routines had stalled for both human scientists and computational approaches.

  Software & Coding · 2023 · solved · 8/10 importance

• ImageNet classification with deep convolutional neural networks

  Accurately classifying 1.2 million high-resolution images into 1000 categories, a task previous methods handled poorly.

  Software & Coding · 2017 · solved · 8/10 importance

• Deep Residual Learning for Image Recognition

  Very deep neural networks were difficult to train, blocking accuracy gains from increased depth in visual recognition.

  Software & Coding · 2016 · solved · 8/10 importance

• Deep Learning-Based Attack Detection and Classification in Android Devices

  Detecting and classifying malware attacks on Android devices, which dominate the global market, remains an ongoing challenge.

  Software & Coding · 2023 · ai assisted · 7/10 importance

• A deep learning‐based approach for software vulnerability detection using code metrics

  Detecting software vulnerabilities traditionally depends on human experts manually defining features, which is tedious and subjective.

  Software & Coding · 2022 · ai assisted · 7/10 importance

• Challenges in Testing Large Language Model-Based Software: A Faceted Taxonomy

  LLM-based software is non-deterministic, so traditional testing cannot verify correctness within individual test cases.

  Software & Coding · 2026 · partially solved · 6/10 importance

• Naive automated machine learning

  Automatically finding the machine learning pipeline with the best generalization performance on a given dataset, a core AutoML task.

  Software & Coding · 2022 · partially solved · 6/10 importance

• HGVul: A Code Vulnerability Detection Method Based on Heterogeneous Source-Level Intermediate Representation

  Detecting vulnerabilities in open-source source code is hard because existing methods lack fine-grained analysis and perform poorly.

  Software & Coding · 2022 · ai assisted · 6/10 importance

• Democratising deep learning for microscopy with ZeroCostDL4Mic

  Novice researchers faced accessibility barriers to training deep learning networks for microscopy due to computational resource and coding requirements.

  Software & Coding · 2021 · partially solved · 6/10 importance

• ICPFuzzer: proprietary communication protocol fuzzing by using machine learning and feedback strategies

  Fuzzing proprietary industrial control system communication protocols is difficult without prior knowledge of their customized, complicated structures.

  Software & Coding · 2021 · partially solved · 6/10 importance

• An Automatic Source Code Vulnerability Detection Approach Based on KELM

  Detecting source code vulnerabilities traditionally requires manually defined rules and features that are costly and perform weakly in practice.

  Software & Coding · 2021 · ai assisted · 6/10 importance

• Dota 2 with Large Scale Deep Reinforcement Learning

  No AI system could defeat world champions at Dota 2, an esports game with long time horizons, imperfect information and complex continuous state-action spaces.

  Software & Coding · 2019 · solved · 6/10 importance

• Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm

  Achieving superhuman play in chess, shogi and Go previously required decades of handcrafted, domain-specific engineering.

  Software & Coding · 2017 · solved · 6/10 importance

• A bilingual benchmark for evaluating large language models

  Lack of standardized benchmarks to evaluate large language models' Arabic capabilities and compare them to English.

  Software & Coding · 2024 · partially solved · 5/10 importance

• Image Denoising: The Deep Learning Revolution and Beyond—A Survey Paper

  Removing additive white Gaussian noise from images, a decades-old problem thought by some to be exhausted before deep learning revived it.

  Software & Coding · 2023 · solved · 5/10 importance

• A Survey of Learning-based Automated Program Repair

  Fixing software bugs automatically to reduce development and maintenance burden.

  Software & Coding · 2023 · ai assisted · 5/10 importance

• A Smart Contract Vulnerability Detection Method Based on Multimodal Feature Fusion and Deep Learning

  Smart contracts on blockchains suffer security vulnerabilities (reentrancy, timestamp dependency, tx.origin, integer overflow) risking substantial user losses, while existing detection methods have low accuracy and typically target only one vulnerability type.

  Software & Coding · 2023 · ai assisted · 5/10 importance

• The integration of machine learning into automated test generation: A systematic mapping study

  Automated generation of software tests is difficult; machine learning may make it effective, but the field's state was unclear.

  Software & Coding · 2023 · partially solved · 5/10 importance

• Competition-level code generation with AlphaCode

  Competitive programming problems require problem-solving skills beyond existing code-generation models; the paper builds a system that generates programs solving such problems.

  Software & Coding · 2022 · partially solved · 5/10 importance

• PyTorch Connectomics: A Scalable and Flexible Segmentation Framework for EM Connectomics

  Reconstructing neurons, synapses, and organelles from nanometer-resolution volumetric microscopy requires scalable segmentation tools that can exploit scarce expert annotations and abundant unlabeled data.

  Software & Coding · 2021 · partially solved · 5/10 importance

• QTune

  Database knob tuning is NP-hard; DBAs cannot tune many instances across environments, and traditional ML methods need hard-to-obtain training data and only support coarse, workload-level tuning.

  Software & Coding · 2019 · ai assisted · 5/10 importance

• COBAYN

  Porting and tuning application code across diverse architectures is slow and costly, and standard compiler optimization levels often fail to deliver the best performance.

  Software & Coding · 2016 · ai assisted · 5/10 importance

• A Review of Large Language Models for Automated Test Case Generation

  Manually creating test cases for software is effortful; the paper reviews whether large language models can automate test case generation.

  Software & Coding · 2025 · ai assisted · 4/10 importance

• <scp>GeoAnalystBench</scp> : A <scp>GeoAI</scp> Benchmark for Assessing Large Language Models for Spatial Analysis Workflow and Code Generation

  It was unclear whether large language models can actually automate real-world geospatial analysis and GIS workflows, and rigorous evaluation of this capability was lacking.

  Software & Coding · 2025 · partially solved · 4/10 importance

• Usage of Large Language Model for Code Generation Tasks: A Review

  The literature lacked an exhaustive analysis of the benefits and drawbacks of using large language models for generating simple and complex code.

  Software & Coding · 2025 · partially solved · 4/10 importance

• COFFE: A Code Efficiency Benchmark for Code Generation

  No suitable benchmark existed to evaluate the time efficiency of LLM-generated code, since correctness benchmarks cannot distinguish efficiency.

  Software & Coding · 2025 · partially solved · 4/10 importance

• Enriching automatic test case generation by extracting relevant test inputs from bug reports

  Automated test generation tools struggle to produce relevant test inputs, limiting the bug-detection effectiveness of generated tests.

  Software & Coding · 2025 · partially solved · 4/10 importance

• Reinforcement learning for mutation operator selection in automated program repair

  Heuristic-based automated program repair wastes resources because mutation operators are chosen uniformly at random, producing many uncompilable or functionality-breaking program variants.

  Software & Coding · 2025 · ai assisted · 4/10 importance

• A Probabilistic Approach to Load Balancing in Multi-Cloud Environments via Machine Learning and Optimization Algorithms

  Efficiently assigning tasks to compatible clouds in dynamic multi-cloud environments is a crucial challenge, especially for ultra-reliable, low-latency applications.

  Software & Coding · 2025 · ai assisted · 4/10 importance

• BioCoder: a benchmark for bioinformatics code generation with large language models

  Large language models lacked a dedicated benchmark for generating bioinformatics-specific code involving domain knowledge, algorithms and data operations.

  Software & Coding · 2024 · ai assisted · 4/10 importance

• SimLLM: Calculating Semantic Similarity in Code Summaries using a Large Language Model-Based Approach

  Existing metrics like BLEU and BERTScore fail to capture semantic nuances and domain-specific terminology when evaluating code summaries.

  Software & Coding · 2024 · partially solved · 4/10 importance

• Controllable Music Playlist Generation Based on Knowledge Graph and Reinforcement Learning

  Conventional music playlist generators struggle to capture a target user's long-term listening preferences.

  Software & Coding · 2022 · ai assisted · 4/10 importance

• Ontology Completion with Graph-Based Machine Learning: A Comprehensive Evaluation

  Growing knowledge bases accumulate wrong or missing entries, and manually finding spurious parts or missing links in ontologies is hard.

  Software & Coding · 2022 · partially solved · 4/10 importance

• Utterance Level Feature Aggregation with Deep Metric Learning for Speech Emotion Recognition

  Automatically recognizing emotions in human speech at the utterance level, needed for applications from mental disease diagnosis to human-robot interaction.

  Software & Coding · 2021 · ai assisted · 4/10 importance

• Leveraging query logs and machine learning for parametric query optimization

  Parametric query optimization must cache a small set of plans for parameterized queries and then quickly pick the best cached plan per query instance.

  Software & Coding · 2021 · solved · 4/10 importance

• Topology-Preserving Deep Image Segmentation

  Segmentation algorithms make topological errors, such as broken connections, on fine-scale structures in images.

  Software & Coding · 2019 · partially solved · 4/10 importance

• Lithuanian Speech Recognition Using Purely Phonetic Deep Learning

  Automatic speech recognition for Lithuanian had attracted little research, and models built for languages like English could not be directly adopted due to complicated language structure and scarce data.

  Software & Coding · 2019 · solved · 4/10 importance

• On Efficiency of Selected Machine Learning Algorithms for Intrusion Detection in Software Defined Networks

  Detecting malicious activity in Software Defined Network data planes requires efficient monitoring methods.

  Software & Coding · 2016 · ai assisted · 4/10 importance

• Clustering-Based Selection for the Exploration of Compiler Optimization Sequences

  Exhaustively exploring all viable compiler optimization sequences is infeasible, yet pass ordering strongly impacts performance and depends on platform and application.

  Software & Coding · 2016 · ai assisted · 4/10 importance

• Automatic feature generation for machine learning--based optimising compilation

  Machine-learned compiler optimisation heuristics depend on hand-chosen program features, and the space of possible features is effectively infinite.

  Software & Coding · 2014 · ai assisted · 4/10 importance

• Continuous learning of compiler heuristics

  Iterative compilation with machine learning finds optimal code optimizations but needs extremely long compilation and training times (days or weeks).

  Software & Coding · 2013 · ai assisted · 4/10 importance

• Using machine learning to partition streaming programs

  Automatically partitioning streaming parallel programs for multicore systems is difficult because the ideal mapping depends on both the program and the underlying architecture.

  Software & Coding · 2013 · ai assisted · 4/10 importance

• Only diff Is Not Enough: Generating Commit Messages Leveraging Reasoning and Action of Large Language Model

  Automatically generating high-quality, informative commit messages for code changes, where prior approaches produced insufficient, poorly readable output.

  Software & Coding · 2024 · ai assisted · 3/10 importance

• Improving Web Element Localization by Using a Large Language Model

  Web-based test automation requires accurately locating web elements, but traditional attribute-matching methods miss context and meaning of elements and words.

  Software & Coding · 2024 · ai assisted · 3/10 importance

• Coordination Between Individual Agents in Multi-Agent Reinforcement Learning

  Existing multi-agent reinforcement learning methods only handle global- or neighborhood-level coordination, leaving coordination between individual agents unsolved in unknown multi-agent environments.

  Software & Coding · 2021 · partially solved · 3/10 importance

• Improving coordination in small-scale multi-agent deep reinforcement learning through memory-driven communication

  Multiple agents trained centrally but executed decentrally struggle to coordinate on tasks requiring synchronisation when they only have partial observations.

  Software & Coding · 2020 · partially solved · 3/10 importance

• Speech Emotion Recognition Using Deep Learning LSTM for Tamil Language

  Automatically recognizing emotions in Tamil speech, which traditional feedforward networks handle poorly.

  Software & Coding · 2021 · ai assisted · 2/10 importance

Search within Software & Coding →