AI News Today

arXiv:2605.00665v2 Announce Type: replace Abstract: The systemic, metabolic, lifestyle factors have established associations with Alzheimer's Disease (AD) through epidemiologic and AD-specific biomarker studies. Whether colored fundus photography (CFP) contains retinal structural signatures corresponding to these AD-related risk domains remains unclear. To determine whether deep learning (DL) models can predict 12 AD-related risk factors from CFP and to characterize the retinal structures underl

arXiv:2605.07821v2 Announce Type: replace Abstract: Out-of-distribution (OOD) detection is crucial for ensuring the reliability of deep learning models. Existing methods mostly focus on regular entangled representations to discriminate in-distribution (ID) and OOD data, neglecting the rich contextual information within images. This issue is particularly challenging for detecting near-OOD, as models with simplicity bias struggle to learn discriminative features in disentangled representations. Th

arXiv:2605.10873v2 Announce Type: replace Abstract: Recovering editable CAD programs from images or 3D observations is central to AI-assisted design, but progress is difficult to measure because existing evaluations are fragmented across datasets, modalities, and metrics. We introduce CADBench, a unified benchmark for multimodal CAD program generation. CADBench contains 18,000 evaluation samples spanning six benchmark families derived from DeepCAD, Fusion 360, ABC, MCB, and Objaverse; five input

arXiv:2606.20005v1 Announce Type: new Abstract: Attention distillation, which trains one attention distribution to match another by minimizing their Kullback-Leibler (KL) divergence, is widely used in knowledge distillation, model compression, continual learning, and sparse-attention LLM training. However, existing approaches materialize both attention distributions before computing the KL reduction, incurring $O(N_QN_K)$ memory and IO costs that become prohibitive at long context lengths. We pr

arXiv:2606.20014v1 Announce Type: new Abstract: Reinforcement learning (RL) has achieved strong performance in sequential decision-making, yet scaling to complex multi-agent environments remains challenging due to sparse rewards, large state-action spaces, and the difficulty of learning coordinated strategies. We propose a hierarchical architecture where a pretrained large language model (LLM) acts as a centralized strategic controller that selects among specialized RL skill policies for a team

arXiv:2606.20104v1 Announce Type: new Abstract: Perception for action suggests that representations of the world should be shaped not by visual fidelity alone, but by their relevance for actions. At the same time, latent JEPA-style world models advocate learning compact predictive states from high-dimensional observations to facilitate the prediction of future states, but end-to-end training of these models is nontrivial because representations may collapse if our only goal is to construct a lat

arXiv:2606.20216v1 Announce Type: new Abstract: Machine learning algorithms deployed for evolving streaming environments must handle the non-stationary data distributions, commonly referred to as concept drift. The presence of concept drift poses a major challenge for many real-world applications because it can severely degrade their predictive performance, hindering their ability to support robust decision-making. Consequently, the timely and efficient detection of drift events is critical for

arXiv:2606.20283v1 Announce Type: new Abstract: Graph neural networks (GNNs) excel at aggregating neighbor information for classification, yet their performance is hindered by graph structural entanglement, where spurious correlations from semantically irrelevant neighbors contaminate node embeddings. This challenge is most acute for nodes near class boundaries in the embedding space, where amplified structural noise blurs decision boundaries and destabilizes predictions. Existing robust GNN met

arXiv:2606.20376v1 Announce Type: new Abstract: Safety is a core concern for deploying reinforcement learning (RL) agents in real-world domains such as robotics and autonomous driving. While benchmarks have been central to progress in RL, existing safety benchmarks with high-fidelity 3D physics remain computationally slow, limiting large-scale experimentation and rapid prototyping. To address this gap, we propose CRAX (Constrained RL Accelerated with JAX). Built on top of the MuJoCo XLA (MJX) ph

arXiv:2606.20474v1 Announce Type: new Abstract: Context-heavy agents place unusual pressure on the key-value (KV) cache: long prefixes are reused across many short turns, while concurrency determines whether the serving system can keep GPUs utilized. We study 4-bit KV-cache compression for this setting, using TurboQuant-style rotation and codebook quantization as a quality anchor and vLLM FP8 KV caching as the deployment anchor. We report three contributions. First, we frame 4-bit KV caching aro

arXiv:2606.20493v1 Announce Type: new Abstract: When large language models serve as evaluators in multi-agent systems, their systematic evaluation biases propagate through the agent network. We introduce Contagion Networks, a formal framework for measuring how evaluator biases spread across interacting LLM agents. In a controlled 3-agent experiment using DeepSeek-chat with three distinct evaluator bias profiles (structured, balanced, evidence-based), we measure the Cross-Agent Contagion Matrix G

arXiv:2606.20560v1 Announce Type: new Abstract: LLM reasoning transparency is a critical affordance for understanding model decisions, mitigating misuse and misalignment, and debugging surprising model behaviors. However, DiffusionGemma performs a larger fraction of its computation in a continuous latent space; does this make its reasoning less transparent? We study this question by decomposing transparency into two components: variable transparency, whether we understand intermediate snapshots

arXiv:2606.19920v1 Announce Type: cross Abstract: Distributed optimization is a highly scalable and structurally transparent technique to solve multi-agent robotics problems; however, such methods often suffer from the need for highly-specialized, problem-specific hyperparameter tunings. In this work, we propose Deep Coordinator, a deep-unfolding framework that learns to dynamically adjust the hyperparameters of ADMM-DDP, a popular distributed solver for robotics tasks, at solve-time in response

arXiv:2606.19947v1 Announce Type: cross Abstract: Reliable quantum control in the presence of decoherence requires policies that combat the effect of environmental noise on the controlled dynamics. Open quantum systems under continuous monitoring generate classical measurement records whose drift depends on the noise experienced by the system; the records of two evolutions sharing the same decoherence channels differ only in this drift, so Girsanov's theorem yields a closed-form, differentiable

arXiv:2606.19989v1 Announce Type: cross Abstract: Modern LLM training breaks a core assumption behind offline batch samplers: the true training cost of a sample is only observable after preprocessing, augmentation, templating, tokenization, and multimodal visual-token expansion. Unless one pays for a preprocessing- and augmentation-dependent length cache, batch construction is therefore blind to the quantity that determines padding, memory use, and GPU saturation. We introduce Online Dynamic Bat

arXiv:2606.20299v1 Announce Type: cross Abstract: Deep learning has managed to evade numerous intuitions from classical statistics to achieve unprecedented performance on a number of real-world tasks. In this article, we investigate the key features and surprises of deep learning from a physics-informed perspective, taking care to point out and justify where possible the many choices inherent in constructing a deep learning model. In particular, we review the phenomenon of neural scaling laws an

arXiv:2606.20041v1 Announce Type: cross Abstract: We propose a model-grounded RAG-based AI economist with an agentic framework for economic scenario analysis using large language models (LLMs) and knowledge graphs. While LLMs can generate fluent economic narratives, economists are often required to make economic claims grounded by economic theory and real-world data. Based on this motivation, this study proposes an RAG-based AI economist, which utilizes knowledge graphs including economic data a

arXiv:2606.20074v1 Announce Type: cross Abstract: Burst suppression (BS) is a clinically relevant electroencephalographic (EEG) pattern used to monitor sedation depth and brain activity in critically ill patients, particularly during induced coma in Intensive Care Units (ICUs). Automatic burst detection remains challenging because BS patterns vary substantially between patients and annotated datasets are scarce. Recently, EEG Foundation Models (FMs) have shown promise across several downstream E

arXiv:2606.20082v1 Announce Type: cross Abstract: The John ellipsoid of a symmetric polytope $P=\{\mathbf{x}\in\mathbb{R}^d:\|\mathbf{A}\mathbf{x}\|_\infty\le1\}$, $\mathbf{A}\in\mathbb{R}^{n\times d}$, is computed by a long line of leverage-score algorithms, from Cohen, Cousins, Lee and Yang (COLT 2019) to its successors [WY24, CLS+25], all reaching a $(1+\varepsilon)$-approximation in $\Theta(\varepsilon^{-1}\log(n/d))$ iterations. We separate this complexity into three costs the modern line c

arXiv:2606.20118v1 Announce Type: cross Abstract: Vision-language-action (VLA) policies have shown strong potential for general-purpose manipulation, yet they often fail on novel, out-of-distribution objects whose appearance or geometry deviates from the training distribution. The standard remedy is to collect multi-view teleoperation data for every failure case, but this scales poorly in both cost and time. We introduce Pose6DAug, a failure-driven data augmentation framework that turns a policy

arXiv:2606.20128v1 Announce Type: cross Abstract: Benchmarks for LLM-generated GPU kernels (KernelBench, TritonBench, GEAK) score correctness through fixed-shape, small-sample allclose-style checks. The number of inputs varies between benchmarks. The shape, dtype, and tolerance are fixed for each kernel. We test that oracle empirically. We construct a controlled corpus of 24 Triton and CPU stand-in kernels (15 correct controls and 9 LLM-style buggy variants seeded with documented transcription e

arXiv:2606.20324v1 Announce Type: cross Abstract: Virtual training environments are software-intensive systems in which reinforcement learning (RL) agents learn, adapt, and demonstrate meaningful behavior. Virtual training environments offer a safe and cost-efficient alternative to training agents in real-world settings. However, to converge, most realistic RL problems require training in multiple, mostly similar but slightly different environments - i.e., families of environment variants. The t

arXiv:2606.20344v1 Announce Type: cross Abstract: Machine learning models have scaled to unprecedented sizes, making training across distributed devices the de facto standard in the field. In this work, we explore how quantum communications can make distributed training both more communication-efficient and information-theoretically private, for both classical and quantum learning models. Ring all-reduce is the foundational communication primitive for large-scale distributed training. We present

Article URL: https://github.com/xyTom/coding-tools-mcp Comments URL: https://news.ycombinator.com/item?id=48594594 Points: 1 # Comments: 0

Hey HN, I built a free tool that lets you sign and upload your iOS project binaries, and serve temporary web pages to sideload on your iPhone straight from Codex Mobile, Claude Code remote control, or any coding agent of choice. No need for VPN or simulator streaming. The web page expires in 30-min but you can access all your builds on our dashboard. Get started with: brew install milq-ai/tap/sqim Comments URL: https://news.ycombinator.com/item?id=48594537 Points: 2 # Comments: 0

<p> 60 second ultrasound-based full-body scanner that beats MRI </p> <p> <a href="https://www.producthunt.com/products/midjourney?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/1175715?app_id=339">Link</a> </p>

Location: Santa Cruz, CA Remote: Yes Willing to relocate: Remote Preferred. SF/Socal Hybrid ok Applied AI / product builder (Full Stack). I’m strongest at turning messy ideas, workflows, datasets, and APIs into useful AI-assisted products and prototypes; especially where product judgment, user understanding, automation, and modern LLM/agent tooling matter. In a world where AI is become exponentially more intelligent in general, a recurring challenge has become harnessing that intelligence for sp

I believe Distribution has an edge Comments URL: https://news.ycombinator.com/item?id=48594425 Points: 2 # Comments: 1

Article URL: https://jeena.net/loop-engineering Comments URL: https://news.ycombinator.com/item?id=48594371 Points: 2 # Comments: 0

We implement an end-to-end workflow for Salesforce CodeGen, loaded from Hugging Face. We move past basic inference by adding function extraction, syntax checking, static safety checks, and unit-test validation. We rerank best-of-N candidates, compose multi-turn program synthesis, and experiment with prompt styles. We finish by visualizing a mini benchmark and exporting the generated artifacts as reusable files. The post Salesforce CodeGen Tutorial: Generate, Validate, and Rerank Python Functions

<p> a small hosted memory layer for AI agents </p> <p> <a href="https://www.producthunt.com/products/pumadb?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/1175700?app_id=339">Link</a> </p>

Article URL: https://www.devclubhouse.com/a/ship-an-mcp-server-in-python-that-exposes-your-internal-api-to-llms Comments URL: https://news.ycombinator.com/item?id=48594116 Points: 1 # Comments: 0

Hi HN, We build agents in Elixir. We kept running into the same issue and found there is no observability for agentic systems. We decided to take the best aspects of LangChain, LangGraph, and DeepAgents and put them into Elixir. BeamWeaver comes with an OTP-native design and: - agents and tool calling - graph workflows - checkpoints and resumable execution - memory stores - retries, fallbacks, interrupts, and human review - typed streaming events - provider adapters for OpenAI, Anthropic, Google

Here is my general take: I feel Gemini is excellent at high-level refactoring but riddled with bugs when writing actual code. On the other hand, GPT/Claude excel at coding, but when it comes to refactoring, they tend to stick to minor patches. They love throwing in unnecessary defensive programming just to maintain backward compatibility, or ending up with verbose spaghetti code. My idea is to complement their strengths: let Gemini provide the directional/architectural ideas, and then have GPT/C

Article URL: https://www.economist.com/leaders/2026/06/18/ai-has-granted-america-vast-new-power Comments URL: https://news.ycombinator.com/item?id=48593869 Points: 2 # Comments: 1

Article URL: https://www.inc.com/bruce-crumley/from-dream-job-to-the-gulag-inside-the-meta-staff-revolt-over-zuckerbergs-brutal-ai-push/91362653 Comments URL: https://news.ycombinator.com/item?id=48593822 Points: 2 # Comments: 0

Article URL: https://techcrunch.com/2026/06/03/metas-ai-agent-for-whatsapp-business-is-now-available-globally/ Comments URL: https://news.ycombinator.com/item?id=48593772 Points: 1 # Comments: 0

A time of turmoil. ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.

Article URL: https://twitter.com/damnventures/status/2067394719756251468 Comments URL: https://news.ycombinator.com/item?id=48593703 Points: 14 # Comments: 0

<pre style='white-space:pre-wrap;width:81ex'>refactor(agents): remove stale file transcript rewrite</pre>