model releases

arXiv:2602.12316v2 Announce Type: replace Abstract: Frontier AI systems are increasingly capable and deployed in high-stakes multi-agent environments. However, existing AI safety benchmarks largely ev

arXiv:2605.23572v1 Announce Type: cross Abstract: In the competitive landscape of sponsored search, balancing retrieval quality with production latency is a critical challenge. While large retrieval m

Hermes Agent now can orchestrate the @OpenHandsDev agents with a new optional skill! `hermes update` then do `hermes skills install official/autonomous-ai-agents/openhands` Reminder: You can already d

arXiv:2605.23821v1 Announce Type: new Abstract: We propose a distributional theory of how hypernymy -- the ``is-a'' relation between general and specific concepts -- is encoded geometrically in langua

arXiv:2506.03530v3 Announce Type: replace-cross Abstract: Multimodal foundation models have demonstrated impressive capabilities across diverse tasks. However, their potential as plug-and-play solutio

arXiv:2605.23628v1 Announce Type: new Abstract: Multi-task benchmarks have become a central pillar of machine learning research, yet their growing influence has incentivised benchmark gaming -- strate

arXiv:2603.10067v2 Announce Type: replace-cross Abstract: Muon has recently shown promising results in LLM training. In this work, we study how to further improve Muon. We argue that Muon's orthogonal

arXiv:2605.22940v1 Announce Type: cross Abstract: Deep learning is increasingly viewed as a dynamical process in parameter space, yet many existing theories still treat training as a closed optimizati

✅Implicit caching is now live on Qwen3.7-Max — kicks in automatically, no setup needed. ⚡️Faster + cheaper out of the box. Need higher, more deterministic hit rates? Try explicit caching instead. 🙌 🔗B

arXiv:2605.22885v1 Announce Type: new Abstract: Formal mathematics libraries are rapidly expanding, creating a growing need to refactor verified proofs for maintainability and to improve training data

Joshua McElwee / Reuters: In his ~43,000-word encyclical, the Pope urged governments to slow down AI development and decried “new forms of slavery” in AI and tech supply chains — Pope Leo urged govern

arXiv:2605.23109v1 Announce Type: new Abstract: AI agents increasingly excel at generating, testing, and refining code. However, they fall short on tasks requiring formal guarantees of full coverage t

arXiv:2605.23187v1 Announce Type: new Abstract: Existing object navigation benchmarks usually tell an embodied agent which object category to find, such as microwave or chair. Human-facing embodied AI

arXiv:2511.03882v2 Announce Type: replace-cross Abstract: Imitation learning-based robot control policies are enjoying renewed interest in video-based robotics. However, it remains unclear whether thi

arXiv:2605.22672v2 Announce Type: replace Abstract: We document inverse scaling in LLMs on forecasting problems whose underlying time series exhibit superlinear growth and tail risk of regime change,

arXiv:2605.23825v1 Announce Type: cross Abstract: It has generally been assumed that geopolitical bias in language models originates from the training data used during the pre-training phase. We teste

I’ve just released MiMo V2.5-Coder. If you have 128 GB of RAM, this is one of the best models you can run locally. It’s fast, and in all my experiments it outperformed Qwen 3.6 and DeepSeek 4-Flash. h

arXiv:2603.24226v3 Announce Type: replace-cross Abstract: Scaling studies for industrial search, advertising, and recommendation have largely emphasized enlarging model capacity or refining architectu

The Claude Code app for iPhone features an interactive Easter egg where scrolling down triggers an animated character named 'Clawd' that jumps and walks around the screen, appearing to search for apps

Just spent a full coding session with Grok Build and honestly? It's right there with Claude. The model is sharp, the agentic flow holds up on complex tasks, and it has actual personality. Few rough ed

arXiv:2605.23082v1 Announce Type: cross Abstract: Survival analysis aims to model how covariates and time jointly shape the time-to-event distribution under right censoring. Classical methods such as

arXiv:2605.22893v1 Announce Type: cross Abstract: We introduce a novel Longitudinal Focused Attention Meditation Electroencephalography (L-FAME) dataset and an accompanying benchmark, designed to fost

arXiv:2603.22348v3 Announce Type: replace Abstract: Online learning algorithms often face a fundamental trilemma: balancing regret guarantees between adversarial and stochastic settings and providing

arXiv:2605.22829v1 Announce Type: cross Abstract: Multimodal Retrieval-Augmented Generation (RAG) has emerged as an effective paradigm for enhancing Large Language Models (LLMs) with external knowledg

arXiv:2605.23203v1 Announce Type: cross Abstract: The adoption of vision neural networks in regulated industries requires formal robustness guarantees, especially in safety-critical domains such as he

arXiv:2601.18685v3 Announce Type: replace-cross Abstract: The capabilities of generative AI in mathematics education are rapidly evolving, posing significant challenges for research to keep pace. Rese

arXiv:2605.23754v1 Announce Type: new Abstract: Developing constitutive models that capture how materials deform under load traditionally requires years of specialized expertise in continuum mechanics

arXiv:2605.23174v1 Announce Type: new Abstract: Remote photoplethysmography (rPPG) enables non-contact measurement of physiological signals from facial videos, offering strong potential for remote hea

arXiv:2605.23007v1 Announce Type: cross Abstract: We explore the application of LLM-driven algorithm optimization to several common tasks in quantitative finance. MadEvolve, a general-purpose algorith

arXiv:2601.14652v5 Announce Type: replace Abstract: While multi-agent systems (MAS) promise elevated intelligence through coordination of agents, current approaches to automatic MAS design under-deliv

arXiv:2605.22872v1 Announce Type: cross Abstract: Experienced physicians develop diagnostic expertise through clinical practice, acquiring not only disease knowledge but also the ability to differenti

arXiv:2602.13480v2 Announce Type: cross Abstract: Launchpads have become the dominant mechanism for issuing memecoins, exposing investors to a new class of high-risk launches that existing rug-pull de

arXiv:2605.22981v1 Announce Type: cross Abstract: Fill-in-the-middle (FIM) is a pretraining objective widely used to equip causal language models with infilling ability, yet its effect on verbatim mem

arXiv:2605.23701v1 Announce Type: new Abstract: We study a protocol-level test for weak-label benchmarks: whether benchmark outputs change when the provided evidence is intervened on. Metadata-only sh

arXiv:2605.23054v1 Announce Type: cross Abstract: Model collapse, the progressive degradation of LLMs trained on their own outputs, has been characterized statistically but lacks a linguistic explanat

arXiv:2605.23057v1 Announce Type: cross Abstract: ModeSwitch-LLM is a lightweight request-boundary controller for improving single-GPU large language model inference efficiency by routing each request

arXiv:2402.14212v4 Announce Type: replace-cross Abstract: Backpropagation's main limitation is its need to store intermediate activations (residuals) during the forward pass, which restricts the depth

arXiv:2605.23871v1 Announce Type: cross Abstract: We develop a gradient flow on the space of probability measures defined on matrix-valued parameters induced by regularized Muon, an analytically smoot

arXiv:2505.17015v2 Announce Type: replace-cross Abstract: Multi-modal large language models (MLLMs) have rapidly advanced in visual tasks, yet their spatial understanding remains limited to single ima

arXiv:2605.23036v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) enable feature-level mechanistic interpretability and activation steering in large language models (LLMs), but SAE-based lang

arXiv:2605.23476v1 Announce Type: new Abstract: Training instabilities in deep networks - loss spikes, oscillatory convergence, and gradient pathologies - are empirically prevalent but lack a rigorous

arXiv:2511.11051v3 Announce Type: replace Abstract: Low-Rank Adaptation (LoRA) fusion enables the composition of subject and style representations for controllable generation without retraining. Howev

arXiv:2605.23652v1 Announce Type: new Abstract: On a 300-persona life-simulation benchmark, pcsp achieves compositional zero-shot persona identification up to 17x above chance, Spearman rho approx 0.7

arXiv:2512.15436v2 Announce Type: replace-cross Abstract: We introduce an extension of the partitioned local depth (PaLD) algorithm that is adapted to online applications such as semi-supervised predi

arXiv:2605.23297v1 Announce Type: new Abstract: AI-enabled services deployed in critical digital infrastructure are subject to governance obligations spanning transparency, accountability, fairness, a

arXiv:2605.23037v1 Announce Type: new Abstract: Data-driven modeling is becoming central to multiphase transport, electronics cooling, acoustic diagnostics, and thermal-fluid digital twins, but progre