Intelligence Digest

OpenAI Blog · 2026-04-09

Explore the official terms and conditions for the OpenAI Full Fan Mode Contest, including eligibility, entry steps, judging criteria, and prize details. Learn how to participate, submit your entry on Instagram, and win IPL match tickets.

OpenAI Blog · 2026-04-08

OpenAI outlines the next phase of enterprise AI, as adoption accelerates across industries with Frontier, ChatGPT Enterprise, Codex, and company-wide AI agents.

OpenAI Blog · 2026-04-08

Discover OpenAI’s Child Safety Blueprint—a roadmap for building AI responsibly with safeguards, age-appropriate design, and collaboration to protect and empower young people online.

OpenAI Blog · 2026-04-06

A pilot program to support independent safety and alignment research and develop the next generation of talent

Hacker News · 2026-04-09

Comments

Hacker News · 2026-04-09

Comments

Hacker News · 2026-04-09

Comments

Hacker News · 2026-04-09

Comments

GitHub Blog · 2026-04-09

In March, we experienced four incidents that resulted in degraded performance across GitHub services. The post GitHub availability report: March 2026 appeared first on The GitHub Blog .

GitHub Blog · 2026-04-08

Get inspired by five of the most memorable, magical, and quirky Universe sessions to date. The post GitHub Universe is back: We want you to take the stage appeared first on The GitHub Blog .

GitHub Blog · 2026-04-06

Discover how Rubber Duck provides a different perspective to GitHub Copilot CLI. The post GitHub Copilot CLI combines model families for a second opinion appeared first on The GitHub Blog .

ScienceDaily — Artificial Intelligence · 2026-04-07

A team of engineers has created a breakthrough memory device that keeps working at temperatures hotter than molten lava, shattering one of electronics’ biggest limits. Built from an unusual stack of ultra-durable materials, the tiny component can store data and perform calculations even at 700°C (1300°F), far beyond what today’s chips can handle. The discovery was partly accidental, but it revealed a powerful new mechanism that prevents heat-induced failure at the atomic level.

ScienceDaily — Artificial Intelligence · 2026-04-06

AI is consuming staggering amounts of energy—already over 10% of U.S. electricity—and the demand is only accelerating. Now, researchers have unveiled a radically more efficient approach that could slash AI energy use by up to 100× while actually improving accuracy. By combining neural networks with human-like symbolic reasoning, their system helps robots think more logically instead of relying on brute-force trial and error.

ScienceDaily — Stem Cells · 2026-04-08

Scientists at Cornell University may be closing in on the long-sought “holy grail” of male contraception: a safe, reversible, nonhormonal method that completely halts sperm production. In a breakthrough mouse study, researchers used a compound called JQ1 to temporarily shut down meiosis—the critical process that produces sperm—without causing lasting harm. After treatment stopped, sperm production bounced back, fertility returned, and the animals produced healthy offspring.

arXiv · 2026-04-08

Large Chunk Test-Time Training (LaCT) has shown strong performance on long-context 3D reconstruction, but its fully plastic inference-time updates remain vulnerable to catastrophic forgetting and overfitting. As a result, LaCT is typically instantiated with a single large chunk spanning the full input sequence, falling short of the broader goal of handling arbitrarily long sequences in a single pass. We propose Elastic Test-Time Training inspired by elastic weight consolidation, that stabilizes LaCT fast-weight updates with a Fisher-weighted elastic prior around a maintained anchor state. The anchor evolves as an exponential moving average of past fast weights to balance stability and plasticity. Based on this updated architecture, we introduce Fast Spatial Memory (FSM), an efficient and scalable model for 4D reconstruction that learns spatiotemporal representations from long observation

arXiv · 2026-04-08

Exact relevance certification asks which coordinates are necessary to determine the optimal action in a coordinate-structured decision problem. The tractable families treated here admit a finite primitive basis, but optimizer-quotient realizability is maximal, so quotient shape alone cannot characterize the frontier. We prove a meta-impossibility theorem for efficiently checkable structural predicates invariant under the theorem-forced closure laws of exact certification. Structural convergence with zero-distortion summaries, quotient entropy bounds, and support-counting arguments explains why those closure laws are canonical. We establish the theorem by constructing same-orbit disagreements for four obstruction families, namely dominant-pair concentration, margin masking, ghost-action concentration, and additive/statewise offset concentration, using action-independent, pair-targeted aff

arXiv · 2026-04-08

Generating motion-controlled videos--where user-specified actions drive physically plausible scene dynamics under freely chosen viewpoints--demands two capabilities: (1) disentangled motion control, allowing users to separately control the object motion and adjust camera viewpoint; and (2) motion causality, ensuring that user-driven actions trigger coherent reactions from other objects rather than merely displacing pixels. Existing methods fall short on both fronts: they entangle camera and object motion into a single tracking signal and treat motion as kinematic displacement without modeling causal relationships between object motion. We introduce MoRight, a unified framework that addresses both limitations through disentangled motion modeling. Object motion is specified in a canonical static-view and transferred to an arbitrary target camera viewpoint via temporal cross-view attention,

arXiv · 2026-04-08

Reservoir engineering has emerged as a powerful paradigm to realize non-reciprocal dynamics in open quantum many-body systems. Here, we show that density-density interactions can transfer bath-induced non-reciprocity between different degrees of freedom. Specifically, we investigate a one-dimensional lattice of spin-$\frac{1}{2}$ fermions with all-to-all Hatsugai-Kohmoto interactions in the presence of an engineered reservoir. We establish the exact solvability of the Lindbladian dynamics and show that the interplay between non-reciprocity and interactions qualitatively reshapes the dynamics of excitations. Remarkably, interactions induce directional drift even in spin sectors that are not directly coupled to the reservoir. By analyzing a driven-dissipative Fermi-Hubbard chain, we show that the same mechanism persists for local interactions. The Hatsugai-Kohmoto model thus emerges as a m

arXiv · 2026-04-08

The rapid growth of generative artificial intelligence (AI) has introduced unprecedented computational demands, driving significant increases in the energy footprint of data centers. However, existing power consumption data is largely proprietary and reported at varying resolutions, creating challenges for estimating whole-facility energy use and planning infrastructure. In this work, we present a methodology that bridges this gap by linking high-resolution workload power measurements to whole-facility energy demand. Using NLR's high-performance computing data center equipped with NVIDIA H100 GPUs, we measure power consumption of AI workloads at 0.1-second resolution for AI training, fine-tuning and inference jobs. Workloads are characterized using MLCommons benchmarks for model training and fine-tuning, and vLLM benchmarks for inference, enabling reproducible and standardized workload p

arXiv · 2026-04-08

Pluralistic alignment has emerged as a critical frontier in the development of Large Language Models (LLMs), with reward models (RMs) serving as a central mechanism for capturing diverse human values. While benchmarks for general response quality are prevalent, evaluating how well reward models account for individual user preferences remains an open challenge. To bridge this gap, we introduce Personalized RewardBench, a novel benchmark designed to rigorously assess reward models' capacity to model personalized preferences. We construct chosen and rejected response pairs based on strict adherence to (or violation of) user-specific rubrics, ensuring that preference distinctions are uniquely tailored to the individual. In particular, human evaluations confirm that the primary discriminative factor between pairs is strictly personal preference, with both responses maintaining high general qu

arXiv · 2026-04-08

Distributed drive electric vehicles offer superior yaw moment control for autonomous drifting in extreme maneuvers. Conventional drift analysis constructs stability boundaries from open loop equilibria points and assumes a fixed envelope structure. However, coupling among control inputs reshapes the phase plane and shifts saddle point location, which can invalidate open loop envelopes when used for closed loop drifting. To address this issue, a saddle point coordinate model is established in this paper by combining a nonlinear tire model with the handling diagram and explicitly accounting for road adhesion coefficient, longitudinal velocity, front wheel steering angle, and additional yaw moment. Based on saddle point properties, an extended dual envelope framework is constructed in the phase plane of slip angle and yaw rate. Using the convergence tendency of state points toward saddle po

arXiv · 2026-04-08

We propose TC-AE, a ViT-based architecture for deep compression autoencoders. Existing methods commonly increase the channel number of latent representations to maintain reconstruction quality under high compression ratios. However, this strategy often leads to latent representation collapse, which degrades generative performance. Instead of relying on increasingly complex architectures or multi-stage training schemes, TC-AE addresses this challenge from the perspective of the token space, the key bridge between pixels and image latents, through two complementary innovations: Firstly, we study token number scaling by adjusting the patch size in ViT under a fixed latent budget, and identify aggressive token-to-latent compression as the key factor that limits effective scaling. To address this issue, we decompose token-to-latent compression into two stages, reducing structural information

arXiv · 2026-04-08

We incorporate an arbitrarily high-order method for the Laplacian operator into the Spectral Difference method (SD). The resulting method is capable of capturing shocks thanks to its a-posteriori limiting methodology, and therefore it is able to survive scenarios in which the dissipative scales (viscous and diffusive) are not properly described. Moreover, it is capable of capturing these scales at lower resolution compared to lower-order methods and therefore attains convergence at lower resolution. We show that the method at hand has exponential convergence when describing smooth solutions and is able to recover a high-order solution when solving the dissipative scales.

arXiv · 2026-04-08

Recent advances in vision-language models (VLMs) have improved image captioning for cultural heritage. However, inferring structured cultural metadata (e.g., creator, origin, period) from visual input remains underexplored. We introduce a multi-category, cross-cultural benchmark for this task and evaluate VLMs using an LLM-as-Judge framework that measures semantic alignment with reference annotations. To assess cultural reasoning, we report exact-match, partial-match, and attribute-level accuracy across cultural regions. Results show that models capture fragmented signals and exhibit substantial performance variation across cultures and metadata types, leading to inconsistent and weakly grounded predictions. These findings highlight the limitations of current VLMs in structured cultural metadata inference beyond visual perception.

NVD · 2023-06-07

The Frontend File Manager plugin for WordPress is vulnerable to Authenticated Settings Change in versions up to, and including, 18.2. This is due to lacking capability checks and a security nonce, all on the wpfm_save_settings AJAX action. This makes it possible for subscriber-level attackers to edit the plugin settings, such as the allowed upload file types. This can lead to remote code execution through other vulnerabilities.

NVD · 2023-06-30

The Web3 – Crypto wallet Login & NFT token gating plugin for WordPress is vulnerable to authentication bypass in versions up to, and including, 2.6.0. This is due to incorrect authentication checking in the 'hidden_form_data' function. This makes it possible for authenticated attackers to log in as any existing user on the site, such as an administrator, if they have access to the username.

NVD · 2023-06-30

The BookIt plugin for WordPress is vulnerable to authentication bypass in versions up to, and including, 2.3.7. This is due to insufficient verification on the user being supplied during booking an appointment through the plugin. This makes it possible for unauthenticated attackers to log in as any existing user on the site, such as an administrator, if they have access to the email.

NVD · 2023-05-25

The MStore API plugin for WordPress is vulnerable to authentication bypass in versions up to, and including, 3.9.2. This is due to insufficient verification on the user being supplied during the add listing REST API request through the plugin. This makes it possible for unauthenticated attackers to log in as any existing user on the site, such as an administrator, if they have access to the user id.

NVD · 2023-05-25

The MStore API plugin for WordPress is vulnerable to authentication bypass in versions up to, and including, 3.9.0. This is due to insufficient verification on the user being supplied during the coupon redemption REST API request through the plugin. This makes it possible for unauthenticated attackers to log in as any existing user on the site, such as an administrator, if they have access to the user id.

NVD · 2023-04-27

The Profile Builder – User Profile & User Registration Forms plugin for WordPress is vulnerable to unauthorized password resets in versions up to, and including 3.9.0. This is due to the plugin using native password reset functionality, with insufficient validation on the password reset function (wppb_front_end_password_recovery). The function uses the plaintext value of a password reset key instead of a hashed value which means it can easily be retrieved and subsequently used. An attacker can leverage CVE-2023-0814, or another vulnerability like SQL Injection in another plugin or theme installed on the site to successfully exploit this vulnerability.

NVD · 2023-04-15

The ZM Ajax Login & Register plugin for WordPress is vulnerable to authentication bypass in versions up to, and including, 2.0.2. This is due to insufficient verification on the user being supplied during a Facebook login through the plugin. This makes it possible for unauthenticated attackers to log in as any existing user on the site, such as an administrator, if they have access to the username.

NVD · 2022-11-10

xterm before 375 allows code execution via font ops, e.g., because an OSC 50 response may have Ctrl-g and therefore lead to command execution within the vi line-editing mode of Zsh. NOTE: font ops are not allowed in the xterm default configurations of some Linux distributions.

NVD · 2022-03-24

The Ninja Forms - File Uploads Extension WordPress plugin is vulnerable to arbitrary file uploads due to insufficient input file type validation found in the ~/includes/ajax/controllers/uploads.php file which can be bypassed making it possible for unauthenticated attackers to upload malicious files that can be used to obtain remote code execution, in versions up to and including 3.3.0

NVD · 2022-06-13

The RSVPMaker plugin for WordPress is vulnerable to unauthenticated SQL Injection due to insufficient escaping and parameterization on user supplied data passed to multiple SQL queries in the ~/rsvpmaker-email.php file. This makes it possible for unauthenticated attackers to steal sensitive information from the database in versions up to, and including, 9.3.2. Please note that this is separate from CVE-2022-1453 & CVE-2022-1505.

NVD · 2023-06-07

The uListing plugin for WordPress is vulnerable to authorization bypass via wp_route due to missing capability checks, and a missing security nonce, in the StmListingSingleLayout::import_new_layout method in versions up to, and including, 1.6.6. This makes it possible for unauthenticated attackers to change any WordPress option in the database.

NVD · 2023-06-07

The Pinterest Automatic plugin for WordPress is vulnerable to authorization bypass due to missing capability checks on the 'wp_pinterest_automatic_parse_request' function and the 'process_form.php' script in versions up to, and including, 1.14.3. This makes it possible for unauthenticated attackers to update arbitrary options on a site that can be used to create new administrative user accounts or redirect unsuspecting site visitors.

NVD · 2023-06-07

The uListing plugin for WordPress is vulnerable to authorization bypass as most actions and endpoints are accessible to unauthenticated users, lack security nonces, and data is seldom validated. This issue exists in versions up to, and including, 1.6.6. This makes it possible for unauthenticated attackers to conduct numerous administrative actions, including those less critical than the explicitly outlined ones in our detection.

NVD · 2023-06-07

The Ultimate Reviews plugin for WordPress is vulnerable to PHP Object Injection in versions up to, and including, 2.1.32 via deserialization of untrusted input in several vulnerable functions. This allows unauthenticated attackers to inject a PHP Object. No POP chain is present in the vulnerable plugin.

Brave Search

The data packages contain both the raw data in their originally published format (CSVs, Excel spreadsheets, and Visual FoxPro database (DBF) files) and metadata that describes how each dataset is partitioned. The PUDL software will download a copy of the appropriate raw inputs automatically as needed and organize them in a local datastore.

Brave Search

A platform for open data of the European power system.

Brave Search

<strong>The platform provides data on installed generation capacity by country/technology, individual power plants (conventional and renewable), and time series data</strong>.

Brave Search

Checking your browser before accessing www.kaggle.com · Click here if you are not automatically redirected after 5 seconds

Brave Search

For each site, we collect 6 hours of the full spectrum scan, the RTL-SDR sweeps the full spectrum from 24 MHz to 1.7 GHz hopping over the frequencies with chunks of 2MHz of bandwidth. Then specific spectrum portions of the licence bands are filtered to label the dataset. The NumPy files then contain PSD data of licensed frequency bands e.g.

Brave Search

Second, the data-driven deep learning solution to classify the radio frequency communication technology used by the transmitter, using transmission features in a compressed space extracted from single PSD measurements over at most 2 MHz band for inference. We release the lightweight version of the framework as Docker container. We release the compressed version of the dataset that contains the measurements of real-world data collected from 47 different sensors deployed across Europe. To access to the dataset: https://zenodo.org/record/7521246

Brave Search

Abstract—While the radio spectrum allocation is well regu- lated, there is little knowledge about its actual utilization over · time and space. This limitation hinders taking effective actions · in various applications including cognitive radios, electrosmog · monitoring, and law enforcement. We introduce Electrosense...

Brave Search

To close this gap, we developed a script called gerber2ems that translates PCB fabrication files (generated for example by KiCad), an open source hardware design suite, into a three-dimensional mesh which can be processed with openEMS. Below, we will review the basic principles of signal integrity analysis with OpenEMS, provide an overview of how the script works and the process of using our open source hardware test board to validate simulation output.

Brave Search

For example, an AC termination might use a 50-ohm resistor in parallel with a capacitor to terminate high-speed clock signals, providing broad frequency range matching (with the cutoff frequency (fc) where the impedance of the termination network matches the impedance of the transmission line is given by the following formula: fc=1/2πRC) Altium Designer Signal Integrity tool enable designers to specify appropriate trace termination · The dielectric constant (Dk) and loss tangent (Df) of PCB materials are critical factors in maintaining signal integrity.

Brave Search

This is especially important to the performance of high-speed digital interfaces, such as MIPI CSI-2, DDR or PCIe. As described on our blog some time ago, basing on the openEMS library, Antmicro created a tool called gerber2ems that converts PCB production files, generated with e.g. KiCad, into PNGs which can then in turn be further processed and used for simulation with openEMS. Signal integrity simulation provides invaluable aid in prototyping complex designs, such as DDR5 memories, high speed FPGAs and SoCs, quicker and more reliably.