Search: “apple”

CUDA: Improve NVFP4 W4A4 activation quantization (#25730) Squash history before conflict-resolution during rebase on master WIP commit Add 32-byte loads, restore per-block amax Use nvfp4x4 intrinsic w

We built this experience based on feedback from early testers and physicians. With your permission, ChatGPT can use relevant context you’ve connected in Health across your conversations — to help you

mtmd : use align_corners for qwen3vl vision position embedding interpolation (#25781) The Qwen3-VL learned position embedding is interpolated to the runtime patch grid with the default bilinear+antial

cuda: GET_ROWS quants (#25962) cuda: add k-quant support to GET_ROWS Device-side embedding lookups require GET_ROWS to handle the k-quants used by common GGUF recipes (Q4_K_M stores token_embd as q6_K

webgpu : add CONV_2D_DW (depthwise conv2d) kernel (#25847) webgpu : add CONV_2D_DW (depthwise conv2d) kernel Implement GGML_OP_CONV_2D_DW for the WebGPU backend, ported from the Vulkan backend's conv2

Most AI SDKs are a wrapper around one engine on one platform. We built the opposite. 7 SDKs. Swift, Kotlin, Flutter, React Native, Web, Electron, rcli. Python. All of them are thin skins over one C++

I've been considering Ollama's Cloud Max plan to replace my GLM Max plan but couldn't find good documentation on how their limits actually translate into GLM 5.2 usage. I know it's by GPU time but tha

Automated lint: 20 errors, 8743 warnings, 3 info

opencl: disable FA and MoE weights repack to work around compiler issues for Adreno 850 GPU (#25745) opencl: workaround for A850 compiler compat opencl: fix DX compiler version parsing and cleanup Co-

[SYCL] Flash Attention with XMX engine via oneDNN (#25222) [SYCL] F16 (default) Flash Attention with XMX engine via oneDNN graph API; Qwen3.6-27b-Q8_0 prefill speed up x1.21 at p=512 and x4.26 at p=80

sycl: Increase minimum buffer size for USM system allocations (#25525) Raise the threshold for minimum buffer size from 1 GiB to 4 GiB, based on real-world experiments of overcommitting device memory

server : refactor prompt cache state ownership (#25649) server : clear checkpoints upon prompt clear server : move the prompt state data to the server_prompt_cache Assisted-by: pi:llama.cpp/Qwen3.6-27

arXiv:2607.06598v1 Announce Type: cross Abstract: Heart rate measurement is one of the key requirements for real-time health monitoring, in particular for health caring of elderly people. Traditional

arXiv:2607.02680v1 Announce Type: cross Abstract: MLLMs have shown strong zero-shot capabilities across diverse inputs such as across images, video, audio, and text. A crucial, yet underexplored, appl

The July 4th weekend All-In @theallinpod turned into a long argument about who owns the intelligence layer. The besties think enterprises just woke up to a trap they had been walking into, here's how

arXiv:2606.28374v1 Announce Type: new Abstract: LLM agents are increasingly improved without weight updates by evolving a natural-language artifact, such as reflections, workflows, playbooks, cheatshe

arXiv:2606.25476v1 Announce Type: new Abstract: Large language models (LLMs) have demonstrated remarkable performance across natural language processing tasks, yet their deployment in high-stakes appl

arXiv:2602.17431v2 Announce Type: replace-cross Abstract: Uncertainty quantification has emerged as an effective approach to closed-book hallucination detection for LLMs, but existing methods are larg

arXiv:2606.10170v1 Announce Type: new Abstract: This paper extends the concept of Learning Entropy (LE) from temporal adaptive systems to spatial learning in multilayer perceptron networks (MLPs) appl

arXiv:2606.07834v1 Announce Type: cross Abstract: LLM judges increasingly turn verdicts into system commitments. Under mixed evidence (claims with both supporting and refuting sources) this is unsafe:

longmemeval experiment arch: 1) deterministic ingestion/extraction (85.6% accuracy, 86.2% retrieval) 2) semantic ingestion/extraction (84.8% accuracy, 94.9% retention) 3) semantic ingestion/determinis

Aleph 2.0 Introducing Runway Aleph 2.0. Edit videos using AI, keeping camera and motion all consistent. Select specific frames in your video to reprompt with GPT Images 2.0 or Nano Banana 2, then appl

arXiv:2509.05614v3 Announce Type: replace-cross Abstract: Pruning is a typical acceleration technique for compute-bound models by removing computation on unimportant values. Recently, it has been appl

arXiv:2605.23747v1 Announce Type: new Abstract: Material segmentation, the pixel-wise classification of physical surface properties, remains a challenging problem in computer vision, requiring physico

arXiv:2605.21671v1 Announce Type: cross Abstract: Hyperspectral super-resolution (HSR) reconstructs a high-spatial-resolution hyperspectral image by fusing a low-resolution hyperspectral image (LR-HSI

arXiv:2605.20584v1 Announce Type: new Abstract: Mobile app marketplaces require developers to disclose standardized content rating descriptors (CRDs) to inform users about potentially sensitive or res

brain dump of how/why we use Evals to measure agents before & after shipping to prod 1. Good Evals simulate what our real users will do and encounter. They’re not really random benchmark tasks, they r

arXiv:2605.18015v1 Announce Type: new Abstract: Production log analytics in self-hosted, resource-constrained environments requires natural-language access to massive log streams without the cost of r

Google LLC is revamping the Google Workspace suite of applications for getting things done, adding new conversational capabilities for brainstorming, an expanded AI Inbox in Gmail, and an all-new appl

arXiv:2605.10117v1 Announce Type: cross Abstract: Autonomous driving scenes range from empty highways to dense intersections with dozens of interacting road users, yet current 3D detection models appl

arXiv:2605.07171v1 Announce Type: new Abstract: The classic multi-armed bandit (MAB) problem tackles the challenge of accruing maximum reward while making decisions under uncertainty. However, in appl

arXiv:2605.03884v1 Announce Type: new Abstract: Multi-agent LLM systems on edge devices need to hand off latent context efficiently, but the practical choices today are expensive re-prefill or full-pr

arXiv:2604.24906v1 Announce Type: cross Abstract: Robotic fruit harvesting often fails to reliably detect whether a fruit has been successfully picked, limiting efficiency and increasing crop damage.

The Ministral-3:3b model requires Ollama 0.13.1, which is in pre-release , and users encountering errors when running it face various issues including memory allocation problems and GPU/CPU offloading

LOCAL AI MODELS ARE CATCHING UP TO FRONTIER MODELS WAY FASTER THAN ANYONE EXPECTED this guy ran qwen 3.6 27B locally on a base macbook pro M4 with 24GB of memory quantized and stripped of safety guard

Bloomberg: How Sergey Brin's political push against a proposed wealth tax in California has helped mobilize a network of fellow tech leaders to sway state issues — In a treehouse nestled in redwoods n

arXiv:2604.21686v1 Announce Type: new Abstract: Interactive video generation models such as Genie, YUME, HY-World, and Matrix-Game are advancing rapidly, yet every model is evaluated on its own benchm

arXiv:2604.19993v1 Announce Type: cross Abstract: Complex-Valued Neural Networks (CVNNs) have significant advantages in handling tasks that involve complex numbers. However, existing CVNNs are unable

Confidently wrong is worse than just wrong. When an AI searches one giant knowledge base for everything, it finds something—and serves it up like it's the answer. That's not a model problem. That's a

arXiv:2604.19966v1 Announce Type: cross Abstract: Vision-language models (VLMs) are increasingly used in settings where sensitivity to low-level image degradations matters, including content moderatio

Kyle Jahner / Bloomberg Law: Music publishers including UMG, Warner Music, and Sony drop a copyright suit against Verizon following a SCOTUS decision limiting ISP liability in Cox's suit — Publishers

arXiv:2604.20156v1 Announce Type: new Abstract: Mixture-of-Experts models, now popular for scaling capacity at fixed inference speed, switch experts at nearly every token. Once a model outgrows availa

Bloomberg: US child safety group NCMEC received 1.5M reports of suspected CSAM with ties to AI in 2025, a significant surge compared to 67,000 in 2024 and 4,700 in 2023 — William Michael Haslach was a

Renju Jose / Reuters: Australia's eSafety Commissioner issues transparency notices to Roblox, Microsoft's Minecraft, and other online gaming platforms to detail child safety measures — Australia's int

Lee Harris / Financial Times: Documents and sources: insurers including QBE and Beazley are moving to cap cyber policy payouts for losses and regulatory fines tied to AI use and “LLMjacking” — Beazley

Jordan Novet / CNBC: LinkedIn names COO Daniel Shapero as its new CEO, succeeding Ryan Roslansky, who will retain his position as EVP at Microsoft — Microsoft has tapped Dan Shapero to be the new CEO