Published , 04:44 Bogota (UTC-5) · 35 sourced items, 18 new since the previous edition · Read the foundations review · RSS
arXiv:2610.24297v1 Announce Type: new Abstract: Until April 2026, the National Vulnerability Database labeled nearly every published Common Vulnerabilities and Exposures (CVE) entry with a Common Vulnerability Scoring System (CVSS) v3.1 vec...
arXiv:2610.25456v1 Announce Type: cross Abstract: Decision models return probabilities over supplied options, but a deployed agent may also need to reply in words. The present study introduces System X, a large language model (LLM) decision...
arXiv:2610.20962v2 Announce Type: replace-cross Abstract: NVIDIA's Grace Hopper (GH200) superchip represents a new generation of heterogeneous computing, tightly integrating a 72-core Arm-based CPU with a Hopper GPU on a single module via a...
What's Changed Fix attention value scaling in MiMo-V2-Flash by @ricky-chaoju in #1971 Fix mask in SMM by @michalk8 in #1979 Muse spark hf names by @nastya236 in #1981 New Contributors @ricky-chaoju made their first contribution in #1971 Ful...
What's Changed fix muse template by @nastya236 in #1977
MLX on Mac with Apple silicon lets AI researchers and developers experiment locally with the latest models and methods in a private environment. The open-source array framework leverages the unified memory architecture of Apple silicon, ena...
What's Changed Adding VJP for the scaled dot product attention. by @tpegolotti in #4563 python: Add pos support to array by @aaishwarymishra in #4488 Add Python 3.15 and 3.15t support by @XXXXRT666 in #4555 Add fixed size unique operation b...
arXiv:2610.09424v3 Announce Type: replace-cross Abstract: Deploying Mixture-of-Experts (MoE) models relies heavily on Expert Parallelism, which generates intense inter-GPU communication. Consequently, state-of-the-art inference systems requ...
[Nightly] Update Swift Package checksums for 2026-10-19 [skip ci]
arXiv:2610.23941v1 Announce Type: new Abstract: Human Activity Recognition (HAR) using wearable or smartphone inertial sensors is commonly evaluated with both classical machine learning and deep learning methods. However, reported performan...
arXiv:2610.23949v1 Announce Type: new Abstract: Collaborative inference serves as a critical methodology in edge intelligence. Existing collaborative inference methods operate by splitting the deep neural network (DNN) model between edge de...
arXiv:2610.24816v1 Announce Type: new Abstract: On-device fine-tuning lets a large language model (LLM) learn from its user's private data without uploading it. Low-rank adaptation (LoRA) reduces the trainable parameters, but each first-ord...
arXiv:2610.25519v1 Announce Type: new Abstract: Test-time scaling (TTS) improves large language model (LLM) reasoning by sampling multiple answers, but fixed sampling budgets ignore both query difficulty and the changing energy cost of infe...
[Nightly] Update Swift Package checksums for 2026-10-18 [skip ci]
Tool: Tree Appreciator Another game experiment with Claude. This one used the default Claude iPhone app (chat mode, not Claude Code mode), where I prompted Claude Opus 5.5 Max with: Build a three js game called "Tree appreciator" - it shows...
sycl: fix compiler warnings - missing prototypes, dead code, extra semicolons ( #30271 ) Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/attestations/57666056 macOS/iOS: macOS Apple Silicon (arm64) macOS Apple...
arXiv:2610.23935v1 Announce Type: new Abstract: Speculative decoding accelerates large language model inference through parallel verification of tokens proposed by a drafter. Diffusion drafters further parallelize proposal generation, but g...
arXiv:2610.23939v1 Announce Type: new Abstract: The size of the KV cache grows linearly with context length, making it a major memory bottleneck in LLM serving. Motivated by recent progress in identifying the functional roles of attention h...
arXiv:2610.24292v1 Announce Type: new Abstract: Learned representations often contain geometric structure induced by symmetries in the underlying data. We ask whether a known continuous parameter space can instead be used to \emph{induce} s...
arXiv:2610.24306v1 Announce Type: new Abstract: Location encoders learn compact representations of latitude--longitude coordinates for downstream geospatial prediction tasks. Existing approaches are pretrained using continuous or naturally ...
server: add --connect, allow p2p access to server outside LAN ( #28277 ) init wip wire it up make it clear that it's experimental add props.connect_code nits fix windows build server: do not advertise a connect code without the tunnel --con...
[Doc][RL] Document the weight-transfer HTTP protocol for trainers wit…
What's Changed openai: allow Responses effort on non-thinking models by @hoyyeva in #19010 llama-server: let llama-server infer the draft speculative type by @dhiltgen in #19064 mlx: add PPLX Decider v1.2 support by @dhiltgen in #19072 Full...
arXiv:2610.26828v1 Announce Type: cross Abstract: Agent learning is shifting from learning from human data toward learning through interaction with the world. However, scaling such learning directly in real-world computer systems is challen...
From code snippets to entire computers, agents have outgrown the sandbox. E2B's next chapter is an agent-driven private cloud that adapts to every task.
Tool: LaTeX renderer Inspired by Paul Ford and Aboard Foundary's new web app Tiny Little Office, which provides a suite of web app content editors (a Word Processor, Outliner, Presentation Builder, Spreadsheet, Calculator, Drawing Tool, Di...
Compare the best 2026 code execution sandboxes for AI agents. See how E2B stacks up against alternatives for isolation, BYOC, billing, and open source.
Redesigning source control at Anthropic Interesting piece from Anthropic describing why and how they built their own custom alternative to Git, based on the dramatic increase in rate of changes to their code after the past six months. In Ma...
Minor Changes 6a74c39 : Require Node.js 22.19.0 or newer and depend on undici@^8 only (undici 7 is dropped). undici 8 respects the server's HTTP/2 SETTINGS_MAX_CONCURRENT_STREAMS and opens more connections when one is saturated. Breaking: t...
arXiv:2610.26124v1 Announce Type: new Abstract: Multi-teacher on-policy distillation (MOPD) aggregates the capabilities of multiple domain-expert teachers into a single student model. However, all domains in MOPD share the student's finite ...
arXiv:2610.26651v1 Announce Type: new Abstract: Pretraining filtering is one of the most common safeguards for open-weight language models. Here we study its effectiveness, durability and scalability using the ESM-2 family of protein langua...
arXiv:2610.23933v1 Announce Type: cross Abstract: Enterprises run schema-guided structured field extraction on documents with frontier multimodal models, and accuracy keeps improving as new models and systems arrive. Two things have not kep...
arXiv:2610.24055v1 Announce Type: cross Abstract: Step-level guards for tool-using agents are evaluated on tidy JSON tool calls, but deployed coding agents act through execute_bash and execute_code tools whose payloads are strings. We intro...
Sources: arXiv API, Apple Machine Learning Research, NVIDIA, Google Research, Google Developers, Microsoft Research, Hugging Face, MLCommons, MIT News, Nature Machine Intelligence, Communications of the ACM, and official GitHub release feeds (MLX, llama.cpp, Ollama, vLLM, MLC LLM, LiteRT-LM, E2B). Items are filtered by topic rules; summaries are AI-assisted (no model; source excerpts shown) and grounded only in each source's own abstract or post text. Always read the linked source before acting.