AX·NEWS
Sign in

Software Engineering

Software Engineering news — Friday 23 October 2026

113 stories from 6 sources, every one in English. Each links to its publisher.

Open in the live feedRSS feedOn a TV

  1. Evolving Threat: http2shell is an RCE in node, Python, and more
    USHacker News Front Page

    Article URL…

  2. Birdwatching rewires the brain and slows cognitive decline
    USHacker News Front Page

    Article URL: https://www.bbc.com/future/article/20261019-birdwatching-rewires-the-brain-and-slows-cognitive-decline Comments URL: https://news.ycombinator.com/item?id=50215997 Points: 469 # Comments: 182

  3. Incremental Computation Graphs
    USlobste.rs

    Comments

  4. CVE Tracker
    USHacker News Front Page

    Article URL: https://cve.0xosprey.com/ Comments URL: https://news.ycombinator.com/item?id=50215530 Points: 201 # Comments: 70

  5. Attention people who think UB is fine
    USlobste.rs

    Comments

  6. Memory is slow, Disk is fast
    USlobste.rs

    Comments

  7. Distributed data, centralized fixes
    USlobste.rs

    Comments

  8. We are all living inside of an obscure sci-fi film from 1970
    USHacker News Front Page

    Article URL: https://thanksforthatjim.bearblog.dev/we-are-all-living-inside-of-an-obscure-sci-fi-film-from-1970/ Comments URL: https://news.ycombinator.com/item?id=50214323 Points: 107 # Comments: 104

  9. Now Google says it might use your Gmail and Google Photos data to target ads
    USHacker News Front Page

    Article URL: https://www.theverge.com/tech/1017600/google-personal-intelligence-privacy-policy-ads-update Comments URL: https://news.ycombinator.com/item?id=50214218 Points: 310 # Comments: 162

  10. Dessa: the most authentic full-duplex speech model ever built
    USHacker News Front Page

    Article URL: https://block.xyz/dessa Comments URL: https://news.ycombinator.com/item?id=50213925 Points: 474 # Comments: 180

  11. Railway Preliminary Report: Oct 21st Security Incident
    USlobste.rs

    Comments

  12. Air Purifiers as White Noise Machines
    USHacker News Front Page

    Article URL: https://www.jefftk.com/p/air-purifiers-as-white-noise-machines Comments URL: https://news.ycombinator.com/item?id=50212507 Points: 145 # Comments: 132

  13. Next-Concept Modeling: Fast, Parallel Coarse-to-Fine Text Generation
    USlobste.rs

    Comments

  14. Can we replace a $150k Looker with a bucket of HTML files?
    USHacker News Front Page

    Article URL: https://david.lequin.dev/blog/bi-software-is-dead/ Comments URL: https://news.ycombinator.com/item?id=50212320 Points: 100 # Comments: 62

  15. A whirlwind tour of systemd-nspawn containers (2025)
    USlobste.rs

    Comments

  16. No Source Code? No Problem! What if you have to port a program, but all you have is a binary? (2003)
    USlobste.rs

    Comments

  17. Kimi K3.1 Pro and K3.1 Pro Highspeed
    USHacker News Front Page

    https://www.kimi.com/code/docs/en/kimi-code/models.html https://artificialanalysis.ai/models/kimi-k3-1-pro https://www.modelcompare.ai/blog/china-open-source-ai Comments URL: https://news.ycombinator.com/item?id=50212065 Points: 1044 #…

  18. UUIDv7 doesn't have the properties you think it does
    USHacker News Front Page

    Article URL: https://alexsci.com/blog/uuidv7-is-weird/ Comments URL: https://news.ycombinator.com/item?id=50211746 Points: 146 # Comments: 117

  19. How is music possible
    USHacker News Front Page

    Article URL: https://www.henrikkarlsson.xyz/p/music Comments URL: https://news.ycombinator.com/item?id=50211719 Points: 132 # Comments: 86

  20. Today, we hacked what we'd set out to hack: the iPhone
    USHacker News Front Page

    Article URL: https://blog.calif.io/p/iphone Comments URL: https://news.ycombinator.com/item?id=50211474 Points: 1046 # Comments: 652

  21. How to Make Things, and Why
    USlobste.rs

    Comments

  22. Release OpenSSL 4.1.0 · openssl/openssl
    USlobste.rs

    Comments

  23. AI-discovered bugs may soon trigger fresh disruptions, cyber experts say
    USlobste.rs

    Comments

  24. eMudhra ignored Mozilla Root investigation deadline; blames social media
    USHacker News Front Page

    Article URL: https://bugzilla.mozilla.org/show_bug.cgi?id=2083449 Comments URL: https://news.ycombinator.com/item?id=50210067 Points: 163 # Comments: 94

  25. Gemma-4-26B-A4B with 39-69% less thinking at same accuracy
    USHacker News Front Page

    Article URL: https://bottlecapai.com/post/thinkingcap-gemma-4-26b/ Comments URL: https://news.ycombinator.com/item?id=50209907 Points: 101 # Comments: 19

  26. Emacs finally folds to my liking: kirigami, tree-sitter, hide-show, Org, XML
    USlobste.rs

    Comments

  27. The glorious @xkcdRandom Zig builtin
    USlobste.rs

    Comments

  28. Colorfigures
    USHacker News Front Page

    Article URL: https://april-zhu.com/colorfigures Comments URL: https://news.ycombinator.com/item?id=50209573 Points: 114 # Comments: 17

  29. Growing Trees
    USHacker News Front Page

    Article URL: https://brashandplucky.com/2026/10/15/growing-trees.html Comments URL: https://news.ycombinator.com/item?id=50209278 Points: 169 # Comments: 24

  30. Profiling Org-＞HTML formal grammar and optimizing it: orgwebalchemy
    USlobste.rs

    Comments

  31. Investor-owned utilities charge 23% more for electricity than public providers
    USHacker News Front Page

    Article URL: https://www.kadoa.com/energy-prices/insights/investor-owned-utilities Comments URL: https://news.ycombinator.com/item?id=50208837 Points: 146 # Comments: 91

  32. a neat trick i found for reusing physical memory
    USlobste.rs

    Comments

  33. Phone Is About to Stop Being Yours: 70 Days Until Lockdown
    USHacker News Front Page

    Article URL: https://keepandroidopen.org/? Comments URL: https://news.ycombinator.com/item?id=50208558 Points: 223 # Comments: 108

  34. Accordion Icons: Which Work Best?
    DESmashing Magazine

    So what icon is the most reliable to show expansion and collapsing for accordions? Is there really any noticeable difference between carets, chevrons, arrows, and triangles? Well, research shows that there is some difference indeed. But…

  35. Software is slow because of bad design, not slow languages
    USlobste.rs

    Comments

  36. OpenSSL 4.1.0
    USHacker News Front Page

    Article URL: https://github.com/openssl/openssl/releases/tag/openssl-4.1.0 Comments URL: https://news.ycombinator.com/item?id=50208099 Points: 167 # Comments: 47

  37. Apple Music 404 Not Found
    USHacker News Front Page

    https://www.macrumors.com/2026/10/23/apple-music-has-a-hidde... Comments URL: https://news.ycombinator.com/item?id=50207910 Points: 906 # Comments: 265

  38. Swedish government offices to ban ChatGPT and Claude
    USHacker News Front Page

    Article URL: https://www.sverigesradio.se/artikel/chat-gpt-och-claude-forbjuder-regeringskansliet Comments URL: https://news.ycombinator.com/item?id=50207405 Points: 194 # Comments: 79

  39. How close can pure FP get to C?
    USlobste.rs

    Comments

  40. Is this the new cyber normal?
    USHacker News Front Page

    Article URL: https://drew.silcock.dev/blog/is-this-the-new-cyber-normal/ Comments URL: https://news.ycombinator.com/item?id=50207115 Points: 168 # Comments: 203

  41. Dutch government had hacked into Jared Kushner's phone
    USHacker News Front Page

    Article URL: https://twitter.com/jaredkushner/status/2113376217415958767 Comments URL: https://news.ycombinator.com/item?id=50206914 Points: 202 # Comments: 132

  42. Job loss and cortisone mania
    USHacker News Front Page

    Article URL: https://tilschuenemann.de/job-loss-cortisone-mania Comments URL: https://news.ycombinator.com/item?id=50206840 Points: 126 # Comments: 61

  43. Color Names
    USlobste.rs

    Comments

  44. What are you doing this weekend?
    USlobste.rs

    Feel free to tell what you plan on doing this weekend and even ask for help or feedback. Please keep in mind it’s more than OK to do nothing at all too!

  45. Disney Raising Prices Days After Confirming Ads Will Now Show Up on Every Tier
    USHacker News Front Page

    Article URL: https://www.tvfandomlounge.com/disney-plus-price-hike-ads-premium-tier-nielsen-emmys-2026/ Comments URL: https://news.ycombinator.com/item?id=50205934 Points: 124 # Comments: 105

  46. Remove pgrust: the entry is cheating
    USHacker News Front Page

    Article URL: https://github.com/ClickHouse/ClickBench/pull/2165 Comments URL: https://news.ycombinator.com/item?id=50205816 Points: 112 # Comments: 52

  47. Debunking "Faster than Rust": Malicious Benchmarks in Tov
    USlobste.rs

    Comments

  48. Rumors of the SaaS demise are greatly exaggerated
    USStack Overflow Blog

    Recorded on the floor of LDX3, Ryan chats with Nivia Henry, an engineering leader most recently at Netflix, about the reality of the supposed SaaS apocalypse, the complex calculus of the modern build-versus-buy decision in the era of AI…

  49. The Many Many Missed Opportunities of Sonic 06
    USlobste.rs

    Comments

  50. The Hibernate Origin Story (Outsider Edition)
    USlobste.rs

    Comments

  51. ConQEC: Concolic Testing for Quantum Error-Correcting Code Implementations
    USarXiv cs.SE (Software)

    arXiv:2610.31443v1 Announce Type: cross Abstract: Quantum error-correcting code (QECC) implementations are hybrid software-control systems: measured syndromes drive decoder dispatch, recovery actions, and logical checks. Existing QECC…

  52. Typed Template Fuzzing
    USarXiv cs.PL (Programming)

    arXiv:2610.31342v1 Announce Type: cross Abstract: Fuzzing software with complex inputs has been a longstanding challenge. Traditionally, fuzzers either programmatically encode language constraints, requiring significant engineering effort…

  53. DynamiQ: Unlocking the Potential of Dynamic Task Allocation in Parallel Fuzzing
    USarXiv cs.SE (Software)

    arXiv:2510.04469v3 Announce Type: replace Abstract: We present DynamiQ, a full-fledged and optimized successor to AFLTeam that supports dynamic and adaptive parallel fuzzing. Unlike most existing approaches that treat individual seeds as…

  54. How Prompt Style Affects AI Accuracy and Speed in Competitive Programming
    USarXiv cs.SE (Software)

    arXiv:2610.30223v1 Announce Type: cross Abstract: Large language models are increasingly used in competitive programming, and their outputs are sensitive to how a prompt is phrased. Prior work has compared prompt strategies for accuracy on…

  55. From Tokens to Credits to Outcomes: Pricing AI Across Models, Applications, and Cloud Platforms
    USarXiv cs.SE (Software)

    arXiv:2610.30701v1 Announce Type: new Abstract: Between 2023 and 2026, AI changed how software is priced. Traditional SaaS charged per seat because serving one more user cost almost nothing. Model inference has non-zero marginal cost, and…

  56. When EVM Defaults Break: Three SDK Patterns for UTXO-Family Wallets
    USarXiv cs.SE (Software)

    arXiv:2610.30183v1 Announce Type: cross Abstract: Mainstream Web3 SDKs (e.g., ethers.js, viem) bundle three coupled assumptions: an account-based ledger, a single secp256k1 signer, and input-selection-free transaction construction. These…

  57. Typed Template Fuzzing
    USarXiv cs.SE (Software)

    arXiv:2610.31342v1 Announce Type: new Abstract: Fuzzing software with complex inputs has been a longstanding challenge. Traditionally, fuzzers either programmatically encode language constraints, requiring significant engineering effort…

  58. Procrustes-Bering-Sheaft: Automated Model Discovery and Virtual Crash Tests for Microservices
    USarXiv cs.SE (Software)

    arXiv:2610.31444v1 Announce Type: new Abstract: Procrustes-Bering-Sheaft automates lightweight model discovery and virtual crash tests for microservices. Each test estimates operation availability by Monte Carlo sampling of service and…

  59. Ohana trees, linear approximation and multi-types for the $\lambda$I-calculus: No variable gets left behind or forgotten!
    USarXiv cs.PL (Programming)

    arXiv:2505.06193v5 Announce Type: replace-cross Abstract: Although the $\lambda$I-calculus is a natural fragment of the $\lambda$-calculus, obtained by forbidding the erasure of arguments, its equational theories did not receive much…

  60. Trace2World: Reconstructing App Worlds from Real Interaction Traces for Scalable GUI Agent Training
    USarXiv cs.SE (Software)

    arXiv:2610.30792v1 Announce Type: cross Abstract: Scaling GUI reinforcement learning requires environments that preserve real-app behavior. Yet coupling behavior inference with coding can turn partial traces into runnable mocks with…

  61. VLTest: Finding Joint Image-Text Failures in Vision-Language Models
    USarXiv cs.SE (Software)

    arXiv:2610.30794v1 Announce Type: new Abstract: Vision-language models (VLMs) are increasingly embedded in software systems. Testing them raises two problems. Mutating either input can change the expected answer, so an output change is not…

  62. Compositional Safety Failures in Harness Evolution: Identification and Runtime Monitoring
    USarXiv cs.SE (Software)

    arXiv:2609.33123v3 Announce Type: replace-cross Abstract: Self-evolving agent harnesses continually update persistent components such as memory, prompts, skills, and tools. We call this process harness evolution. However, such evolution…

  63. Detection Is Not Enough: Reasoning Traces as Testable Program Hypotheses
    USarXiv cs.SE (Software)

    arXiv:2610.30200v1 Announce Type: new Abstract: When a language model is deployed, no reference solution exists, so the reasoning trace it writes is the only evidence for judging its answer. The reference-free substitutes each fail…

  64. CAMGen: A Core Change-Aware Multi-View Framework for Commit Message Generation
    USarXiv cs.SE (Software)

    arXiv:2610.31552v1 Announce Type: new Abstract: Commit messages help developers understand code changes and support software maintenance. Recent approaches use repository context to improve commit message generation (CMG) with large…

  65. Adaptive Skill Granularity for Coding Agents: Runtime Routing and Macro-to-Micro Recovery
    USarXiv cs.SE (Software)

    arXiv:2610.31461v1 Announce Type: cross Abstract: Reusable skills allow large language model agents to package task-specific procedures, tool affordances, and execution guidance into modular units. Prior work has largely studied how agents…

  66. Stale Instruction Files Override the Code in Coding Agents: A One-Sentence Fix
    USarXiv cs.SE (Software)

    arXiv:2610.30571v1 Announce Type: new Abstract: Purpose: Coding agents follow a project's instruction file, such as CLAUDE.md or AGENTS.md, even when it disagrees with the existing code and even when they notice the disagreement. We…

  67. What the Trace Shows and What the Label Says: Auditing Behavioral Detectors in Agent Evaluation
    USarXiv cs.SE (Software)

    arXiv:2610.30491v1 Announce Type: cross Abstract: LLM agents are often evaluated by reducing execution traces to labels such as final-answer correctness, tool verification, and fault propagation. These labels are measurements produced by…

  68. MCP vs CLI for AI Agents: A Comparative Analysis of Tool Integration Strategies for Autonomous Systems
    USarXiv cs.SE (Software)

    arXiv:2610.30157v1 Announce Type: cross Abstract: AI agents require interfaces to interact with external systems such as files, APIs, databases, and cloud services. Two competing paradigms have emerged: CLI (Command Line Interface), where…

  69. Can Loan Processors and Underwriters Spot AI-Edited Statements? An Empirical Study
    USarXiv cs.SE (Software)

    arXiv:2610.30758v1 Announce Type: cross Abstract: AI-assisted editing can produce loan documents that preserve familiar issuer layouts while changing the applicant's identity. We investigated whether experienced underwriters and loan…

  70. The Reversible Intermediate Language L4, its Type System, and its Bidirectional Small-Step Semantics
    USarXiv cs.PL (Programming)

    arXiv:2610.31201v1 Announce Type: new Abstract: L4 is a generic, reversible, linearly-typed, low-level language intended as an intermediate language for compiling reversible programming languages to reversible machines. L4 is intended to…

  71. A Multi-Dimensional, Per-Pass Empirical Study of the LLVM Optimization Pipeline
    USarXiv cs.SE (Software)

    arXiv:2606.31238v2 Announce Type: replace Abstract: Quantifying the marginal impact of individual optimization passes underpins phase ordering, pass selection, optimization design, and analysis of pass/hardware interactions. In LLVM, the…

  72. APO-DD: Dialogue Disentanglement via Reflective Optimization of Multi-Component Prompts
    USarXiv cs.SE (Software)

    arXiv:2606.07894v2 Announce Type: replace Abstract: In multi-party chat, interleaved dialogues make it difficult to identify the context of each utterance. Dialogue disentanglement separates these dialogues to support dialogue analysis and…

  73. Agentic Verified Optimization of Real-World Software
    USarXiv cs.PL (Programming)

    arXiv:2610.30188v1 Announce Type: cross Abstract: Program optimization aims to improve performance while preserving behavior. Coding agents can already make programs faster by changing algorithms, data structures, and execution strategies…

  74. Structure-Aware Equivalence for Evaluating Autoformalized Statements
    USarXiv cs.PL (Programming)

    arXiv:2610.25018v3 Announce Type: replace Abstract: Evaluating autoformalized mathematical statements requires determining whether a candidate preserves the mathematical claim intended by a trusted reference. Existing methods approach this…

  75. A Multi-Dimensional, Per-Pass Empirical Study of the LLVM Optimization Pipeline
    USarXiv cs.PL (Programming)

    arXiv:2606.31238v2 Announce Type: replace-cross Abstract: Quantifying the marginal impact of individual optimization passes underpins phase ordering, pass selection, optimization design, and analysis of pass/hardware interactions. In LLVM…

  76. Agentic Verified Optimization of Real-World Software
    USarXiv cs.SE (Software)

    arXiv:2610.30188v1 Announce Type: new Abstract: Program optimization aims to improve performance while preserving behavior. Coding agents can already make programs faster by changing algorithms, data structures, and execution strategies…

  77. JANUS: Agentic Cross-Language Fuzzing Harness Generation for Android Native Libraries
    USarXiv cs.SE (Software)

    arXiv:2610.30218v1 Announce Type: new Abstract: Modern Android apps extensively rely on native libraries, bridging Java/Kotlin and C/C++ through the Java Native Interface (JNI). This cross-language interoperation, however, exposes the…

  78. Agentic Architecture of Information Systems. The AAM/5C Model: A Reference Model for Distributed Autonomy
    USarXiv cs.SE (Software)

    arXiv:2610.30221v1 Announce Type: new Abstract: Large organizations increasingly face what this book calls the cognitive complexity wall: at the scale of hundreds of interacting components, maintaining the dependency map and reacting at…

  79. The Customer Is the Builder: How AI Coding Agents Are Rewriting the Economics of Software
    USarXiv cs.SE (Software)

    arXiv:2610.30307v1 Announce Type: new Abstract: AI coding agents could change whether organizations buy software or build the functionality they need. We focus on organizations and professional buyers whose domain knowledge can help them…

  80. What Ethereum Developers Ask For, and Whether They Get It: Ten Years of Ethereum Stack Exchange
    USarXiv cs.SE (Software)

    arXiv:2610.30829v1 Announce Type: new Abstract: The Ethereum community learns where developers struggle through surveys and interviews, which record what developers say when asked about a chosen theme. Ethereum Stack Exchange (ESE) records…

  81. ATLAS: A Multi-Agent Framework for Lifting Binary to Canonical Compiler IR
    USarXiv cs.SE (Software)

    arXiv:2610.30877v1 Announce Type: new Abstract: Lifting stripped, heavily optimized binaries to a canonical, pre-optimization LLVM intermediate representation (IR) is essential for downstream security analysis and legacy software migration…

  82. When Code Gets Cheap, Evaluation Gets Scarce: An Experience Report on AI-Assisted Agile Software Engineering
    USarXiv cs.SE (Software)

    arXiv:2610.30916v1 Announce Type: new Abstract: Coding assistants are widely reported to speed up code production; most accounts stop there. We report the end-to-end construction and deployment of a compliance-critical, human-in-the-loop…

  83. Superintelligent Software Engineering: Governance That Scales with Capability
    USarXiv cs.SE (Software)

    arXiv:2610.31306v1 Announce Type: new Abstract: Software agents are being trained to exceed human engineers; we argue that capability alone does not produce better software engineering. We define superintelligent software engineering as…

  84. Automating Feature Flag Cleanup at Scale with Multi-Agent LLM Systems: An Industrial Experience at DoorDash
    USarXiv cs.SE (Software)

    arXiv:2610.31351v1 Announce Type: new Abstract: Feature flags are runtime configuration mechanisms used for gradual feature rollouts, experimentation, and kill switches. Once a feature is fully rolled out, the flag and its gated code…

  85. Can LLMs Improve Their Own Harness? A Controlled Study on ARC-AGI-3
    USarXiv cs.SE (Software)

    arXiv:2610.30499v1 Announce Type: cross Abstract: A generally intelligent system acquires new skills efficiently without requiring humans to redesign its software scaffold for every new task. Yet language models rarely act alone. A harness…

  86. Model of Things (MoT): A Model-Driven Approach to Streamline Cloud-of-Things Application Development
    USarXiv cs.SE (Software)

    arXiv:2512.14613v2 Announce Type: replace Abstract: The convergence of Cloud Computing and the Internet of Things (IoT) underpins modern, scalable, and data-intensive systems. Cloud-of-Things (CoT) applications are now pervasive across…

  87. LIDL: LLM Integration Defect Localization via Knowledge Graph-Enhanced Multi-Agent Analysis
    USarXiv cs.SE (Software)

    arXiv:2601.05539v2 Announce Type: replace Abstract: LLM-integrated software, which embeds or interacts with large language models (LLMs) as functional components, exhibits probabilistic and context-dependent behaviors that fundamentally…

  88. Measuring the Permission Gate: A Stress-Test Evaluation of Claude Code's Auto Mode
    USarXiv cs.SE (Software)

    arXiv:2604.04978v3 Announce Type: replace Abstract: Claude Code's auto mode is the first deployed permission system for AI coding agents, using a two-stage transcript classifier to gate dangerous tool calls. Anthropic reports a 0.4% false…

  89. No Edges, No Verdict: A Large-Scale Empirical Study of Declared Dependency Graphs in 78K SBOMs in the Wild
    USarXiv cs.SE (Software)

    arXiv:2607.22140v3 Announce Type: replace Abstract: Software Bills of Materials (SBOMs) are consumed not only as component inventories but as dependency graphs: vulnerability triage, reachability filtering, and impact analysis all traverse…

  90. PosterHarness: Turning Scientific Poster Generation into an Auditable Instruction-Following Benchmark
    USarXiv cs.SE (Software)

    arXiv:2607.03006v2 Announce Type: replace-cross Abstract: Text-rich image models can now design poster-scale layouts, but we lack ways to measure whether they honor scientific communication contracts: legible labels, prescribed aspect…

  91. Operator Packages, Proposer Strength, and Construction-Family Plateaus in Office-Scale Verified Search
    USarXiv cs.SE (Software)

    arXiv:2609.29636v2 Announce Type: replace-cross Abstract: Verified search, in which a language model proposes programs, a hard evaluator scores them, and selection keeps the best, has recently moved mathematical records; controlled…

  92. Deep Search in Real-World Web Environments
    USarXiv cs.SE (Software)

    arXiv:2610.30181v1 Announce Type: cross Abstract: Despite substantial progress in Deep Search agents, their ability to conduct end-to-end research in real-world open Web environments remains insufficiently evaluated. Unlike static…

  93. TrustedCode: Evaluating Large Language Models for Developing Trusted Code through Automated Program Verification
    USarXiv cs.PL (Programming)

    arXiv:2610.30432v1 Announce Type: cross Abstract: Recent reports, such as "Who is using AI to code?", indicate that approximately 30\% of Python code contributions to GitHub are AI-generated. Given the inherent unreliability of this code…

  94. Beyond Accuracy: Disentangling Decomposition from Typed-State Quality Assurance in Multi-Agent LLM Pipelines
    USarXiv cs.SE (Software)

    arXiv:2610.31280v1 Announce Type: new Abstract: Multi-agent LLM pipelines are often credited with gains in software engineering tasks, but it is rarely clear whether a gain stems from the typed contract between agents or from the…

  95. Governance-State Corruption in Long-Term Memory for LLM Agents: A Cross-Model Stress Test
    USarXiv cs.SE (Software)

    arXiv:2610.30698v1 Announce Type: cross Abstract: An LLM agent can retrieve a memory correctly yet use it incorrectly after its validity state changes. We study governance-state corruption: faults that alter which retained records enter…

  96. From Prose to Properties: Testing Research Claims with ARATest for Agent-Native Science
    USarXiv cs.SE (Software)

    arXiv:2610.30495v1 Announce Type: cross Abstract: AI agents now run experiments and report findings, so research produces claims faster than researchers can check them. Each empirical claim implies a property that must hold over its cases…

  97. ProductWebBench: Evaluating Website Continuity Beyond Task Completion
    USarXiv cs.SE (Software)

    arXiv:2610.31118v1 Announce Type: new Abstract: Frontend coding agents are usually graded on one question: did the requested target get built. Real engineering changes carry a second obligation, because the rest of the site has to keep…

  98. VERA: A Responsible AI Dashboard for Assessing LLM Deployments in Light of the EU AI Act
    USarXiv cs.SE (Software)

    arXiv:2610.31588v1 Announce Type: new Abstract: Institutions that deploy large language models in regulated settings often need technical evidence of responsible AI practices before approving a deployment. COMPL-AI provides a technical…

  99. A Consolidated Deployment Edition of a Microservice Core-Banking Platform: Composing Services by Reference into Three Hosts
    USarXiv cs.SE (Software)

    arXiv:2610.30212v1 Announce Type: new Abstract: Microservices give teams independent deployment and scaling, but each deployed service also reserves compute and adds operational work. This paper examines a second deployment edition of a…

  100. REGATTA: A Cross-Domain Measurement Framework and Fault-Injection Harness for Enterprise AI Agents
    USarXiv cs.SE (Software)

    arXiv:2610.30924v1 Announce Type: new Abstract: Enterprise AI-agent evaluation needs evidence about whether test suites expose specified faults and whether their reference rules are trustworthy. We introduce REGATTA (Regulatory Evaluation…

  101. Extending FunctionGemma for Practical On-Device Mobile Function Calling
    USarXiv cs.SE (Software)

    arXiv:2609.25373v2 Announce Type: replace-cross Abstract: On-device assistants require function-calling models that map natural language to local system actions, but existing resources emphasize web APIs or narrow mobile-action catalogs…

  102. Finitely Verifiable Segmentation: Decidability for Every Tokenizer with a Finite-State Verifier
    USarXiv cs.PL (Programming)

    arXiv:2610.30168v1 Announce Type: cross Abstract: A segmentation policy assigns to each byte string at most one segmentation into tokens. Call it finitely verifiable when a finite automaton over marked strings, bytes paired with boundary…

  103. Separate Field Scoring Protects Unused Tools When Agents Pool Tool-Use Experience
    USarXiv cs.SE (Software)

    arXiv:2610.30867v1 Announce Type: new Abstract: Agents that call tools from a shared catalog can pool what they learn about which tool serves which request. We measure what pooled usage examples do to tool retrieval, reporting separately…

  104. Code Comprehensor: Hierarchical Documentation and Cross-Modal Retrieval for Large Repositories
    USarXiv cs.SE (Software)

    arXiv:2610.30016v2 Announce Type: replace Abstract: Understanding large, complex, and rapidly evolving code bases remains a core challenge for software engineering at scale, hindering onboarding, maintenance, refactoring, and impact…

  105. Self-Evolving Static Analysis via LLM Pseudo-Labeling
    USarXiv cs.SE (Software)

    arXiv:2610.30184v1 Announce Type: cross Abstract: Security bugs often slip past static analysis not because they are impossible to detect, but because detection rules have not yet been written for the code in front of us. New frameworks…

  106. TrustedCode: Evaluating Large Language Models for Developing Trusted Code through Automated Program Verification
    USarXiv cs.SE (Software)

    arXiv:2610.30432v1 Announce Type: new Abstract: Recent reports, such as "Who is using AI to code?", indicate that approximately 30\% of Python code contributions to GitHub are AI-generated. Given the inherent unreliability of this code…

  107. AuditCoder: Process-Auditable Code Generation through Responsibility-Preserving Task Graphs
    USarXiv cs.SE (Software)

    arXiv:2607.29529v2 Announce Type: replace Abstract: Plans can guide code generation without record?ing which code implements each subproblem, leaving reviewers to reconstruct these links dur?ing handoff and maintenance. We present…

  108. Tracing Hallucinations in LLM Reasoning for UML Class Diagram Generation: An Empirical Taxonomy and Mitigation Study
    USarXiv cs.SE (Software)

    arXiv:2610.30946v1 Announce Type: new Abstract: Large Language Models (LLMs) can generate UML class diagrams from natural language requirements, but hallucinations may compromise the correctness and completeness of the resulting diagrams…

  109. From Natural Language to Verified C: Verification-Guided Generation with ACSL Specifications
    USarXiv cs.SE (Software)

    arXiv:2610.30725v1 Announce Type: new Abstract: AI-assisted code generation amplifies the demand for automated formal verification. Generated programs must not only be executable but also accompanied by machine-checkable evidence of…

  110. This I Believe (1952)
    USHacker News Front Page

    Article URL: https://www.awakin.org/v2/read/view.php?tid=468 Comments URL: https://news.ycombinator.com/item?id=50203759 Points: 139 # Comments: 62

  111. UUIDv7 doesn't have the properties you think it does
    USlobste.rs

    Comments

  112. The foolishness of "natural language programming" (1978)
    USHacker News Front Page

    Article URL: https://www.cs.utexas.edu/~EWD/transcriptions/EWD06xx/EWD667.html Comments URL: https://news.ycombinator.com/item?id=50203161 Points: 331 # Comments: 222

  113. Please don't mess with links
    USlobste.rs

    Comments