AI News Today

<pre style='white-space:pre-wrap;width:81ex'>refactor(agents): extract attempt prompt phase</pre>

<p>There are currently 1,000+ open Quantum Information Scientist positions globally.</p> <p>For context: there are only about 50 Quantum Cryptographer roles. There are 200 Quantum Algorithms Researcher roles. But 1,000 Quantum Information Scientist positions.</p> <p>This role is bigger than most people realize. And the salary range — $110K-$180K — reflects the fact that organizations understand how critical this work is.</p> <h2> What A Quantum Information Scientist Actually Does </h2> <p>You're

<p>There are 7,000 open Quantum Software Engineer positions right now. But most people don't get hired for these roles because they're chasing the wrong things.</p> <p>They think they need a physics degree. They don't.<br> They think they need quantum certifications. They don't.<br> They think they need to understand all of quantum mechanics. They really don't.</p> <p>What they actually need is a path. A clear, practical sequence of steps from wherever they are now to being hired.</p> <p>Here ar

<pre style='white-space:pre-wrap;width:81ex'>refactor(agents): remove dead test surfaces (#107113)</pre>

<p>Most people think building a support chatbot is about generating human-like text. It isn’t. The real problem is memory. Or more specifically, the complete lack of it.</p> <p>I got tired of prompt engineering hacks and started looking for something better. I didn’t want a bot that sounded smart for one message and then immediately forgot everything like it hit its head on a table.</p> <p>That’s how IRIS (Intelligent Recall & Issue Support) started.</p> <p>Instead of building yet another chatbo

<p>Gemma 4 is arriving in model catalogs with an unusually broad capability surface. Google's <a href="https://ai.google.dev/gemma/docs/core" rel="noopener noreferrer">official overview</a>, last updated July 8, 2026, lists five sizes, text and image input across the family, audio on selected variants, native system-role support, and 128K or 256K context windows depending on model size.</p> <p>Those are model-family properties. They do <strong>not</strong> prove that a particular hosted endpoint

Article URL: https://github.com/provos/ironcurtain Comments URL: https://news.ycombinator.com/item?id=48902269 Points: 1 # Comments: 0

<pre style='white-space:pre-wrap;width:81ex'>fix(plugin-sdk): preserve runtime contexts in test mocks (#107059) * test(plugin-sdk): make runtime context mock stateful * docs(agents): record Testbox invocation contracts</pre>

<p>Hey everyone,</p> <p>I wanted to share a small technical detail from working on my content sites. When I started integrating AI-assisted writing, the biggest challenge wasn't generating content, but making it sound human. Initial drafts often felt flat, predictable, and frankly, a bit robotic.</p> <p>My solution involved a few post-processing steps. First, a 'burstiness' pass. This isn't about randomizing, but analyzing sentence length variance and adjusting it to mimic human writing patterns

Article URL: https://eshumarneedi.com/2026/07/13/handson-with-ios-and-ipados.html Comments URL: https://news.ycombinator.com/item?id=48902247 Points: 1 # Comments: 0

<pre style='white-space:pre-wrap;width:81ex'>refactor(ai): remove redundant tool argument finalization</pre>

<p><strong>The Model Context Protocol is the connective tissue between LLMs and external systems — filesystems, databases, APIs, browsers, internal tooling.</strong> In a well-functioning agentic deployment, MCP lets a model read documents, query databases, and take actions in the world. In an adversarial deployment, MCP becomes the mechanism by which attacker-controlled content reaches a model's context window and convinces it to take actions the user never authorized.</p> <p>We previously cove

<p>There's a quiet crisis happening in quantum computing that nobody wants to talk about.</p> <p>Quantum startups are desperate to hire mathematicians. They're posting job descriptions that look like they were written by a physics textbook. "Required: PhD in quantum physics, expertise in topological error correction, 5+ years of quantum algorithm development." Then they complain that the talent pool is tiny.</p> <p>Of course the talent pool is tiny. They're looking for unicorns who don't exist.<

<pre style='white-space:pre-wrap;width:81ex'>fix(onboarding): harden live inference handoff (#107041) * fix(onboarding): harden live inference handoff * fix(crestodian): preserve auto runtime probes * fix(ci): restore current main lint gates</pre>

<p>The AI agent hype is real. Everyone wants to build autonomous systems that can write code, make decisions, and operate unsupervised. But here's the problem: most companies hiring for agent development are using the same outdated vetting criteria they used for regular software engineers.</p> <p>And they're building garbage.</p> <p>I've spent the last decade recruiting deep tech talent. Over the past year, I've watched the agent market explode. And I can tell you with absolute certainty: the bo

<pre style='white-space:pre-wrap;width:81ex'>fix: restore grouped AI auth options in guided onboarding (#107038) Restore the grouped provider/auth picker for fresh guided onboarding, keep secondary providers under More, expose OAuth and device-code methods, and bound the setup probe session identifier so ChatGPT login is not rejected by the prompt cache key limit.</pre>

<p><em>Originally published at <a href="https://devopsdiary.blog/posts/validators-judge-they-dont-help/" rel="noopener noreferrer">devopsdiary.blog</a>. Post F-AID3 in the "Governing AI in the Enterprise" series.</em></p> <p>A validator that fixes the thing it just flagged has quit its job. It stopped grading and started doing the homework, in the same motion, and now you can't tell which part of the result you're supposed to trust.</p> <p>I keep running into this in the tooling everyone's shipp

<p> AI sales platform for modern B2B teams </p> <p> <a href="https://www.producthunt.com/products/pebbles-ai?utm_campaign=producthunt-atom-posts-feed&utm_medium=rss-feed&utm_source=producthunt-atom-posts-feed">Discussion</a> | <a href="https://www.producthunt.com/r/p/1195845?app_id=339">Link</a> </p>

arXiv:2607.09664v1 Announce Type: new Abstract: To provide a structured and interpretable assessment, we decompose the image-based diagnosis into components following the Toulmin model of argumentation. This model consists of a claim, grounds, warrant, qualifier, rebuttal, and backing. Consider a claim generated by a machine learning (ML) model for retinal diagnosis. Rather than accepting this claim at face value, one could either apply explainable AI (XAI) methods or adopt an argumentation-base

arXiv:2607.09665v1 Announce Type: new Abstract: Prompt wrappers often differ only in formatting, yet they can change model scores enough to flip leaderboard conclusions. We study this variance under a token-controlled protocol and introduce two complementary metrics: the Format Sensitivity Index (FSI), the accuracy range induced by wrapper choice, and the Parseability Sensitivity Index (PSI), the corresponding range in answer parseability. Across 140,000 OpenRouter generations spanning 7 QA task

arXiv:2607.09678v1 Announce Type: new Abstract: When LLM agents hand off information to one another, does the message format matter? Two literatures disagree: format-optimization work reports that structured messages cut cost without hurting accuracy, while format-restriction work finds that imposing structure degrades generation -- and neither measures what happens when a message traverses multiple hops, where copy fidelity, not one-shot generation, dominates. We introduce a controlled relay te

arXiv:2607.09706v1 Announce Type: new Abstract: Language models turn a worded situation into a numeric plan, and the dominant pipelines (NL4Opt, OptiMUS, ORLM, OR-LLM-Agent) commit to a single objective and point-valued coefficients, then solve once. For decisions that allocate real budget, effort, or clinical attention, that confidence is the failure mode: every objectified number is an assumption, and a plan optimal only if the guesses are exactly right is fragile -- mimicry of computation. YU

arXiv:2607.09709v1 Announce Type: new Abstract: Post-training a code generator against a learned judge can optimize proxy features that raise the score without improving the artifact. We study the opposite signal: a deterministic, judge-free, ungameable filter -- whether a generated project launches cleanly under a headless engine (strict-launch). Under this gate, rejection-sampling self-distillation compounds out-of-family generalization. On GameCraft-Bench (mapping a natural-language brief to

arXiv:2607.09713v1 Announce Type: new Abstract: A key step toward autonomous industrial operation is the ability to create and reconfigure control policies from natural-language requirement specifications, with minimal or no manual redesign. In this setting, policy generation by AI agents can be a credible path when paired with a plant-aware validator (e.g., a digital twin) that can check generated candidate actions before execution. However, practical deployment is constrained by inference late

arXiv:2607.09714v1 Announce Type: new Abstract: The Feedback-Coupled Memory Systems (FCMS) architecture formalizes closed-loop coordination through four abstract operators, two of which - the agent update operator $f_i$ and the environmental update operator $\Psi$ - are left axiomatically undefined in the original framework. To address this, $f_i$ is defined by Mechanism-Based Intelligence (MBI), where agents update locally through a decentralized price mechanism and economic principles, and $\P

arXiv:2607.09739v1 Announce Type: new Abstract: We study LLM benchmark coreset selection: selecting a small subset of prompts over multiple benchmarks whose induced model scores and rankings approximate those obtained from the full benchmark suite. In evaluation-unsupervised benchmark coreset selection (our approach), the selection algorithm uses no model evaluation outcomes, and operates on a fine granularity by producing subsets of prompts over multiple benchmarks rather than producing a sub-c

arXiv:2607.09740v1 Announce Type: new Abstract: Safe motion planning in advanced driver-assistance systems and autonomous vehicles requires an accurate understanding of how the surrounding traffic scene is likely to evolve. However, many existing lane-change prediction methods remain centered on a single target vehicle, while multi-agent forecasting approaches often describe scene evolution only through future positions and provide limited explicit information about the maneuver associated with

arXiv:2607.09743v1 Announce Type: new Abstract: We investigate whether structured reasoning interventions improve the strategic economic reasoning of large language models, and whether their effects depend on model architecture. Using Hotelling's linear city model as a diagnostic vehicle, we evaluate GPT-4.1-mini (a standard instruction-following model) and GPT-5-mini (a reasoning-optimized model) under five conditions - an unscaffolded baseline and four reasoning interventions - across eight qu

arXiv:2607.09744v1 Announce Type: new Abstract: Least privilege, the principle that an identity should hold only the permissions strictly required for its task, has been a foundational primitive of access control for decades. We argue that this principle is insufficient for agentic AI systems, which do not merely hold permissions but can combine, approve, and amplify them across workflows and system boundaries. We propose least autonomy as an appropriate generalization and develop a formal theor

arXiv:2607.09748v1 Announce Type: new Abstract: In distributed systems, the classical State Machine Replication (SMR) model assumes that correct replicas execute deterministic transitions to yield identical bitwise states. However, the rise of agentic distributed systems -- where autonomous, stochastic, and model-driven agents orchestrate infrastructure -- presents scenarios where deterministic, bitwise replication is insufficient. Replicas operating with generative models may exhibit divergent

arXiv:2607.09751v1 Announce Type: new Abstract: Machine Learning (ML) algorithms have been widely used to estimate agricultural variables across diverse contexts. However, because the quantity and quality of training data strongly influence performance of ML algorithms, their use can be constrained by limited or incomplete reference data. Synthetic Data Generation (SDG) offers a practical approach to address this issue by producing artificial but realistic samples that preserve key characteristi

arXiv:2607.09762v1 Announce Type: new Abstract: Public battery aging datasets are a critical asset for advanced health management, but their practical use is often limited by inconsistent formats, unclear schemas, and metadata scattered across repositories and publications. Current curation remains largely manual and hard to reproduce, while general-purpose data integration tools miss the domain-specific semantics of electrochemical time-series data. We present BatteryLake, a governed data lakeh

arXiv:2607.09765v1 Announce Type: new Abstract: A cheap swarm of unreliable agents can be steered to a correct consensus by a few strong, expensive "oracle" correctors. We ask how much one must spend, and where to place the oracles. We model the swarm as a consensus on a graph in which each oracle pins one node toward the truth at a cost-coupled, concave strength, and measure quality by the coherence H(R)=tr M(R)^{-1}. Our first result is that H stays submodular (each added oracle helps less tha

arXiv:2607.09766v1 Announce Type: new Abstract: AI agents are increasingly deployed in shared environments where they pursue diverse goals and compete for rewards. This multi-agent competition can lead to behaviors that serve individual gains at collective cost -- for instance, marketing agents may post misleading content as a result of competing for engagement on social media. Human societies address such problems through norms that constrain acceptable behavior, supported by enforcement mechan

arXiv:2607.09770v1 Announce Type: new Abstract: Industrial agentic AI systems increasingly exhibit a gap between prototype capability and production deployment. In particular, adaptive agents may generate plausible outputs while remaining difficult to verify under non-determinism, confidentiality constraints, limited context, and weak observability. This paper formulates a bounded verification protocol for adaptive agentic controllers represented by finite symbolic rules, explicit diagnostic pre

arXiv:2607.09773v1 Announce Type: new Abstract: Computer-use agents must solve long-horizon tasks through repeated interaction with partially observable, multimodal desktop environments. Although imitation learning and offline trajectory refinement provide strong priors, static traces cannot cover the causal feedback loop of real computer use: each action changes the screen state, future action space, and recovery options. EvoCUA-1.5 extends self-evolving computer-use agents from offline experie

arXiv:2607.09786v1 Announce Type: new Abstract: Length-penalized reinforcement learning can shorten chain-of-thought reasoning while hiding an influence that drives the model's answer. In our experiments, training with length penalties does not stop misleading hints from steering models, even though the models' chains of thought mention the hint much less often. A token-accuracy evaluation would count these runs as successful because they use fewer reasoning tokens with little accuracy loss; it

arXiv:2607.09789v1 Announce Type: new Abstract: We introduce PHITSBench, an execution-scored benchmark for the Monte Carlo Particle and Heavy Ion Transport code System (PHITS). PHITSBench comprises 282 transport-scorable tasks spanning three common workflow categories: parameter editing (Edit), syntax repair (Repair ), and complete simulation generation from natural-language descriptions (Reproduce). Each task is evaluated using a Composite Metric Score that combines execution success with agree

arXiv:2607.09768v1 Announce Type: new Abstract: In this paper, we present the first (to the best of our knowledge) demonstration of a low-power MCU-based edge device for Automatic License Plate Recognition (ALPR). The design leverages on a 9-core RISC-V processor, GAP8, coupled with a QVGA ultra-low-power greyscale imager. The proposed visual processing pipeline uses a multi-model inference approach based on SSDlite-MobilenetV2 for license plate detection and LPRNet for optical character recogni

arXiv:2607.09779v1 Announce Type: new Abstract: The deep nonnegative matrix factorization (DNMF) technique is proposed to address the low interpretability of deep learning-based methods in extracting multilayer features from synthetic aperture radar (SAR) target samples. However, existing DNMF methods employ a layer-by-layer decomposition strategy, which is prone to causing error accumulation and local optimum, thereby hindering a consistent improvement in recognition accuracy as the number of l