How Does Machine Learning Manage Complexity?
arXiv:2604.07233v1 Announce Type: new Abstract: We provide a computational complexity lens to understand the power of machine learning models, particularly their ability to model complex systems. Mach
Knowledge catalogue
arXiv:2604.07233v1 Announce Type: new Abstract: We provide a computational complexity lens to understand the power of machine learning models, particularly their ability to model complex systems. Mach
arXiv:2604.07709v1 Announce Type: cross Abstract: Ask a frontier model how to taper six milligrams of alprazolam (psychiatrist retired, ten days of pills left, abrupt cessation causes seizures) and it
arXiv:2503.18018v2 Announce Type: replace Abstract: We demonstrate that large language models' (LLMs) mathematical reasoning is culturally sensitive: testing 14 models from Anthropic, OpenAI, Google,
arXiv:2509.22981v2 Announce Type: replace Abstract: We study a class of multi-stage stochastic programs, which incorporate modeling features from Markov decision processes (MDPs). This class includes
arXiv:2604.08171v1 Announce Type: new Abstract: Accurate ocean mapping is essential for applications such as bathymetry estimation, seabed characterization, marine litter detection, and ecosystem moni
People in Washington get played, yet again We really should worry about cybersecurity - a lot – but Mythos is not the model these guys think it is. (See my newsletter today for three reasons why it is
maybe some nuance 😄 I don’t think anyone is “lying” about how great Mythos will be —> but there’s expectation misalignment between the Test Harness set up for Mythos and a belief it was given this cra
Want more proof that Anthropic's PR has no idea what it's talking about? The talk of Mythos being 'their most aligned model ever'. They could perhaps truthfully speak about 'new high scores on our ali
Wrote up some thoughts on Anthropic's Project Glassing, where their latest Opus-beating model is available to partnered security research organizations only Given recent alarm bells raised by credible
So I created this simple test (for myself) to see how good the new models are in analytic output. This is not asking them to draw a SVG (which just tests how much SVG or HTML they saw) but creating a
model : Granite switch decision 256 options (#30693) feat(convert): Attempt to come up with 256 single-token, semantic option markers Each switch model needs to identify a collection of single-token o
arXiv:2610.22321v1 Announce Type: cross Abstract: As joint audio-video generation models become more capable, they can generate audio-visual scenes exhibiting three forms of complexity: multi-event co
model : MTP support for Mellum2.1 (#30612) model : MTP support for Mellum model : inline rope construction in Mellum Website: https://llama.app Attestations: https://github.com/ggml-org/llama.cpp/atte
arXiv:2610.22292v1 Announce Type: cross Abstract: Large language models (LLMs) are becoming prevalent in scenarios where honesty is critical. However, despite alignment training, these models still ex
arXiv:2610.21741v1 Announce Type: new Abstract: Closed-loop evaluation is widely regarded as a more faithful way to assess end-to-end (E2E) driving models than open-loop evaluation, and Bench2Drive, b
arXiv:2610.22245v1 Announce Type: new Abstract: Small language models are increasingly deployed on edge devices and robotic systems to meet cost and latency constraints, yet their limited capabilities
arXiv:2610.21483v1 Announce Type: new Abstract: Large language models can produce fluent answers that the supplied context does not support. A compact frozen proxy model can detect such answers by rea
arXiv:2610.20690v1 Announce Type: cross Abstract: The present work is concerned with how to measure, at scale, the multi-dimensional impact of generative AI models on the individuals who use them. In
Mistral AI announced in mid‑October 2026 that its new model, Mistral Large 4, achieved high scores on Ulam’s private STEM benchmark (STEMBench), particularly excelling in engineering and astronomy cat
arXiv:2610.21555v1 Announce Type: cross Abstract: Muon's gains in data efficiency motivate evaluating its performance under explicit data, time, and tuning budgets. On a single A100 GPU, we compare a
arXiv:2412.12048v2 Announce Type: replace Abstract: Low-Rank Adaptation (LoRA) has become a standard approach for parameter-efficient fine-tuning of image diffusion models, with over 200K models train
arXiv:2604.03925v2 Announce Type: replace Abstract: Large language models (LLMs) struggle to accumulate evidence across multiple rounds of user interactions, failing to update their beliefs in a manne
arXiv:2610.19704v1 Announce Type: new Abstract: Large language models (LLMs) increasingly help users arrange appointments, plan journeys, and allocate limited resources. These tasks require models to
arXiv:2610.14346v2 Announce Type: replace Abstract: Enterprise customer-support agents increasingly rely on Large Language Models (LLMs) to execute bounded, tool-using workflows. We study this problem
arXiv:2610.18010v2 Announce Type: replace-cross Abstract: We introduce Fish Audio Drama 3, an open-source speech generation model that turns text into directed performances, built on a 35B-parameter m
arXiv:2610.17757v1 Announce Type: new Abstract: Large language model agents can follow complex human instructions to solve real-world tasks. In practice, however, initial instructions are often incomp
arXiv:2610.14576v1 Announce Type: new Abstract: As frontier LLMs gain more capabilities, the cost to run them also goes up. At the same time, small language models (SLMs) also gain more capabilities b
Meta Platforms Inc. today released open weights for its Muse Spark artificial intelligence model, which had been proprietary since its April debut. The release, called Muse Spark Open, is built for co
NVIDIA’s open‑source Nemotron 3.5 Super VL is a 120 B parameter mixture‑of‑experts multimodal model that activates only 12 B parameters per token, enabling combined reasoning, instruction following, t
Two months ago we released a fully quantized NVFP4 Qwen3.8-27B trained with QUASAR. We've since trained QUASAR checkpoints for a bunch of other recent models: Qwen3.5-4B, Gemma 4 E4B and 12B, and Muse
arXiv:2610.15513v1 Announce Type: cross Abstract: Safety-aligned large language models (LLMs) remain vulnerable to adversarial prompts that bypass their safety training. Existing automated red-teaming
arXiv:2610.17665v2 Announce Type: new Abstract: Can vision-language models turn 2D observations into 3D understanding that supports precise action? This requires inferring object geometry, anticipatin
arXiv:2610.19232v1 Announce Type: new Abstract: Large language models impose substantial memory demands and arithmetic costs during inference. Dense-to-mixture-of-experts (MoE) conversion is a model o
arXiv:2610.20456v1 Announce Type: new Abstract: The scaling relationship of compute and performance for language models (LMs) has been extensively studied; more recently, models are being designed to
arXiv:2610.17528v1 Announce Type: cross Abstract: Training recipes are tuned by hand for one model, cluster and workload, and every new model, accelerator, framework version or dataset means tuning ag
arXiv:2610.14605v1 Announce Type: cross Abstract: Current evaluations of multimodal unlearning typically focus on whether a model stops outputting target information. This surface-level criterion, how
arXiv:2610.15565v1 Announce Type: new Abstract: Recent multimodal large language models achieve promising results on visual reasoning tasks, approaching human-level performance on existing STEM benchm
Elon Musk’s artificial intelligence company SpaceXSI, formerly xAI, today released Grok Voice 3.0, a new version of the speech-to-speech model behind its Voice Agent API. SpaceXSI is pitching the mode
arXiv:2610.16784v1 Announce Type: new Abstract: Generating and editing presentation slides requires models to combine accurate text rendering, instruction following, visual grounding, aesthetic design
This model is a 74B parameter, 9B-active-parameter mixture-of-experts (MoE) model, packed into 24GB using 2-bit quantization with advanced quantization-aware training techniques to retain quality. It
arXiv:2610.18836v1 Announce Type: new Abstract: Vision-Language Models (VLMs) have achieved strong performance in visual perception and spatial reasoning, leading to growing interest in using them as
arXiv:2610.14519v1 Announce Type: new Abstract: Effectively eliciting the capabilities of large language models (LLMs) at inference time remains an important challenge. Existing studies address this c
arXiv:2610.14921v1 Announce Type: new Abstract: A language model that keeps learning online must not slowly erode its own base capabilities. We ask which part of a modern LLM should absorb continual u
arXiv:2610.15660v1 Announce Type: new Abstract: Domain generalised semantic segmentation (DGSS) has recently benefited from combining multiple foundation models, such as vision-language and vision fou
Very cool to see Embedding Gemma 2 by @Google, number on trending on HF! The media and silicon valley bubble only talks about reasoning LLMs when there is a much bigger variety of AI models that are s
What AI Safety needs is better tooling and engineering. For instance, building stronger red-teaming tools. Great to see this open-source effort to red-team image models. It seems to learn better ways
arXiv:2610.17844v1 Announce Type: new Abstract: Spatial understanding is a necessary step in moving multimodal large language models (MLLMs) toward reasoning about the 3D physical world. Their visual
When we talk about agents, that doesn't necessarily mean we're talking about Python. In today's modern ecosystem, an agent might run as a high-concurrency Go microservice on Cloud Run, or live inside
Imagine your compliance team notices that payments are taking longer than expected to clear. They suspect fraud, and you’re the data scientist helping figure out which transactions to investigate firs
model: support vllm decision 2 model (#30378) models: support pplx-decider vllm d2 update clarify naming rule stricter validation fix ci rm redundant filter_tensors Website: https://llama.app Attestat
model: support vllm d3 models (#30383) models: support pplx-decider vllm d2 update clarify naming rule stricter validation d3 Website: https://llama.app Attestations: https://github.com/ggml-org/llama
arXiv:2502.01025v4 Announce Type: replace Abstract: Large language models (LLMs) process entire input contexts indiscriminately, which is inefficient when the information required to answer a query is
Release: llm-meta-ai 0.2 Meta AI release Muse Wonder (and Muse Wonder Full), their new 'most capable models'. These look like they may be Meta's first Fable class models - Artificial Analysis scores M
arXiv:2610.12881v1 Announce Type: new Abstract: Understanding and manipulating spatial relations is fundamental to AI systems that reason about physical environments and construct grounded world model
arXiv:2610.13790v1 Announce Type: cross Abstract: How well can vision-language models (VLMs) reason about what is missing from an image? We study this question with OccludeBench, a manually curated be
Strata vs Infernix on the same model, RTX 5090 — A/B at 262k and 512k context Hi folks, another inference engine post for your feed. I was blown away by Strata, but saw someone else post about Inferni
arXiv:2603.01376v2 Announce Type: replace Abstract: Sparse plus Low-Rank (mathbf{S} + mathbf{LR}) decomposition of Large Language Models (LLMs) has emerged as a promising direction in model compressio
arXiv:2610.11112v1 Announce Type: cross Abstract: Image-generation models can now produce text-rich, natural-looking visual artifacts that are hard to distinguish from real-world evidence, such as new
arXiv:2610.11373v1 Announce Type: cross Abstract: Continual learning remains challenging for large language models, which must enable models to acquire new skills and knowledge without degrading exist
arXiv:2610.11064v1 Announce Type: cross Abstract: A language model (LM) can usually answer the same question in more than one way, but reinforcement learning with verifiable rewards (RLVR) is indiffer