Knowledge catalogue
arXiv:2610.05977v1 Announce Type: new Abstract: Serving a large language model (LLM) across a fleet of deployments requires several weight-precision operating points. Multi-precision formats serve the
arXiv:2610.04467v1 Announce Type: new Abstract: Large language models (LLMs) remain highly vulnerable to jailbreak attacks that induce harmful behaviors and circumvent safety alignment. To defend agai
arXiv:2610.05025v1 Announce Type: new Abstract: Dual-system Vision-Language-Action (VLA) models improve real-time robotic control by pairing a slow, reasoning-capable generalist with a fast specialist
arXiv:2610.04426v1 Announce Type: cross Abstract: Machine unlearning seeks to remove the influence of designated training data from a trained model without retraining from scratch. Retrain-free method
arXiv:2610.04303v1 Announce Type: new Abstract: Recursive computation repeatedly compresses or reuses intermediate states, creating a simple tension: information that must remain useful across longer
arXiv:2610.05322v1 Announce Type: new Abstract: Test-time compute can improve mathematical reasoning, but can short-budget runs predict how mathematical reasoning scales with additional compute? We in
arXiv:2610.04845v1 Announce Type: new Abstract: Aligning large language models (LLMs) with human values is important for safe, efficient, and beneficial AI deployment. However, human values are multif
arXiv:2610.05341v1 Announce Type: new Abstract: Chronic wound management affects over 8.2 million patients in the United States and imposes substantial clinical and economic burden. Clinical wound ass
arXiv:2603.19617v2 Announce Type: replace Abstract: Federated learning (FL) is a communication-efficient framework for solving distributed optimization problems. Standard FL methods aggregate local up
arXiv:2610.03102v1 Announce Type: cross Abstract: An LLM agent can recognize uncertainty yet still choose the wrong next step: asking when action is already justified, or seeking clarification when th
arXiv:2610.02775v1 Announce Type: new Abstract: Mental health stigma has profoundly harmful impacts but its complexity makes it difficult to evaluate. Stigma may involve explicit derogation, but also
kv-cache: fix restoring mismatched KV cache rotation by saving exact rotation metadata (#28498) kv-cache: save exact KV rotation metadata, reject restoring mismatched rotation tests : move the state r
arXiv:2610.03141v1 Announce Type: new Abstract: Video multimodal large language models (MLLMs) keep climbing video question answering benchmarks, yet shuffling the frames, masking the segment that sup
arXiv:2610.03615v1 Announce Type: new Abstract: Recent advances in frontier models enable robot manipulation from only a few demonstrations, but high inference latency limits their use for real-time r
arXiv:2610.02711v1 Announce Type: cross Abstract: Human activity recognition (HAR) from wrist-worn accelerometers is increasingly used for health and behavioral tracking. Yet, most wearable HAR models
arXiv:2610.02654v1 Announce Type: new Abstract: Models of social contagion usually assume how individuals adopt beliefs and derive population behavior from it. We instead empirically measure belief ad
arXiv:2610.03314v1 Announce Type: cross Abstract: Flow- and diffusion-based generative models have recently emerged as flexible and highly efficient forecasting models for dynamical systems. When comb
arXiv:2610.02494v1 Announce Type: new Abstract: Automatic horizon tracking is a foundational task in 3D seismic interpretation. Most existing deep learning approaches formulate it as dense semantic se
arXiv:2610.03491v1 Announce Type: new Abstract: Most long-term time-series forecasting models map the look-back window directly to the full horizon in a single pass. While efficient, this design does
arXiv:2610.03020v1 Announce Type: new Abstract: Long-term agents must remember not only what is true about a user, but also how a particular agent should work with that user as their shared history ev
arXiv:2610.02951v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures scale language models efficiently by activating only a few experts per token, but the saving is confined to com
arXiv:2610.03474v1 Announce Type: new Abstract: Federated learning (FL) enables collaborative training of medical image segmentation models without sharing raw patient data, yet existing approaches as
Great overview of context compression in LLM Agents And one interesting, unexpected finding. If your compaction policy is tuned to cut tokens, it may be making your agent run slower. Great study from
arXiv:2610.02241v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) architectures allow frontier language models to scale to trillions of parameters, but their deployment is constrained by mass
arXiv:2610.03063v1 Announce Type: new Abstract: Large Language Models (LLMs) are prone to generating hallucinated content, which compromises their reliability in knowledge-intensive tasks. To address
See how Arize uses Harbor and Phoenix to benchmark AI agents, MCP servers, skills, and CLIs in reproducible sandboxes, then inspect traces and annotations. The post How we benchmark AI agents and tool
arXiv:2610.03574v1 Announce Type: new Abstract: We introduce HyperBrowseComp, a multilingual and multimodal browsing benchmark comprising 423 manually authored and human-validated questions across 13
arXiv:2610.02571v1 Announce Type: cross Abstract: As AI-assisted programming becomes increasingly mainstream, the environmental impact of AI-generated software has emerged as an important consideratio
GLM 5.3 from Z.ai is now available on Amazon Bedrock: a 753B-parameter mixture-of-experts model built for coding and long-horizon agentic tasks. Learn how to invoke it with the OpenAI-compatible APIs,
arXiv:2610.02670v1 Announce Type: cross Abstract: LLM agents are known to be slow in rollouts. An agent completes a task one step at a time. At each step, it reasons and then chooses an action to exec
arXiv:2610.03702v1 Announce Type: new Abstract: The choice of post-training data for large language models substantially affects downstream performance. Gradient-based data selection is a popular appr
Let's go https://huggingface.co/reflection! Introducing Beam: a highly efficient agentic open model with 501B total parameters and 23B active. - Frontier reasoning efficiency - Advances the Western op
arXiv:2610.02418v1 Announce Type: cross Abstract: Modern large language models (LLMs) are trained on massive, largely unfiltered datasets, including content scraped from nearly every accessible websit
arXiv:2610.02827v1 Announce Type: new Abstract: Modern AI systems are designed to refuse hazardous requests. A jailbreak is a prompt crafted to bypass those safeguards and elicit outputs that the syst
arXiv:2610.03056v1 Announce Type: new Abstract: Large language models are increasingly used as reasoning components in AI-driven materials Co-Scientists, yet the reliability of the resulting verificat
arXiv:2606.18717v2 Announce Type: replace-cross Abstract: Turkish is agglutinative: meaning is carried by morphemes, yet the subword tokenizers that drive modern language models split words by corpus
arXiv:2610.03201v1 Announce Type: cross Abstract: Gaussian Processes (GPs) are a powerful tool for modelling and quantifying uncertainty in functional relationships. However, they require practitioner
arXiv:2605.06639v2 Announce Type: replace-cross Abstract: We introduce Recursive Agent Optimization (RAO), a reinforcement learning approach for training recursive agents: agents that can spawn and de
arXiv:2610.02976v1 Announce Type: new Abstract: Audio-Visual Large Language Models (AV-LLMs) remain prone to cross-modal hallucinations, where one modality incorrectly affects predictions about anothe
arXiv:2610.02718v1 Announce Type: cross Abstract: Vision-language models such as CLIP exhibit strong semantic generalization, but remain limited in fine-grained visual perception. A recent work named
arXiv:2610.03165v1 Announce Type: new Abstract: Several variance-reduced versions of REINFORCE based on importance sampling achieve an improved O(epsilon^{-3}) sample complexity to find an epsilon-sta
arXiv:2610.02702v1 Announce Type: new Abstract: Multi-agent debate is increasingly used to reach consensus among LLM agents, yet agents often yield to a unanimous majority. When an agent changes its a
GitHub (llama.cpp fork) : https://github.com/sky-is-green/prism-ml-llama.cpp (Prebuilt binaries there to try instantly) Draft model : https://huggingface.co/SkyIsNotGreen/Scion-35B-A3B-mtp-drafter sub
arXiv:2610.03000v1 Announce Type: cross Abstract: Intrinsic explainability remains a challenging problem, particularly in contexts where multilayer perceptrons (MLPs) require dynamic re-training withi
arXiv:2610.03124v1 Announce Type: cross Abstract: Open-weight language models can be downloaded, modified, and deployed beyond their developers' control, limiting the effectiveness of centrally enforc
The new v0.6.0 release packs a lot of good stuff: - Clef support (text + vision) - High-quality support for Qwen3.8-Flash-Next - Massive performance improvement with Metal - New `llama_batch_ext` API
arXiv:2610.02399v1 Announce Type: new Abstract: Multimodal agents are increasingly used for data visualization tasks but remain limited in autonomous review. Unlike humans, they may fail to recognize
arXiv:2610.02713v1 Announce Type: new Abstract: Most KV-cache compression methods classify attention heads once, either offline or during prefill, and keep this classification fixed throughout generat
arXiv:2610.02614v1 Announce Type: new Abstract: AI models serving a heterogeneous population must act on the principles appropriate to each user and context. While post-training has been shown to narr
common : prepare load_from_models_dir() for path conversion (#29674) This is part of the fs::path modernization series. That was also the opportunity to remove fs_list(). Signed-off-by: Adrien Gallouë
So Kimi K2 is outdated, and so is GPT OSS 120b. Which of the modern open weights models can boast the least sycophancy? I need this both for creative/research assistant usage (sycophancy led me down b
Not a bot. Not a Strata shill. Just sharing my experience. So, I have an R9700 in my machine, plus an RTX 5060 Ti, and 64GB DDR5 system RAM. Overall, not a bad setup. Anyway, I mainly run a daily driv
What's Changed Models run on MLX on Apple Silicon by default In this release, on Apple Silicon devices, model architectures supported by the MLX runtime will automatically run on MLX. ollama pull qwen
Krea 2 Turbo — 2-Step Distillation LoRA (work in progress) Previous posts/releases - here, here and here. 🧪 A fast-preview adapter, from a project still in training. When the subject is close and fill
Language is not fundamental to human consciousness. In terms of evolution, it arrived only recently on our timeline. Consciousness evolved over millions of years through sensory awareness, emotion, an
Hi folks, it's Lewis here from the post-training team at Hugging Face. We've been exploring how to train open models in different coding harnesses and wrote up a looong guide on how we solved this usi
arXiv:2610.00111v1 Announce Type: new Abstract: Model judges now supervise multimodal systems at scale, filtering training data, selecting outputs, and supplying the reward that shapes multimodal reas
Claude Desktop on Amazon Bedrock is limited to the model's knowledge cutoff without web search. In this post, we walk through connecting Claude Desktop to Web Search using Amazon Bedrock AgentCore Gat
arXiv:2610.00465v1 Announce Type: cross Abstract: Next-generation large language models (LLMs) are expanding from the cloud to ubiquitous edge devices. However, edge devices typically either lack the
arXiv:2610.00854v1 Announce Type: new Abstract: Frontier vision-language models (VLMs) combine scene estimation, interaction grounding, and executable actions. Understanding how these abilities suppor