Search: “model-releases”

arXiv:2604.13287v1 Announce Type: new Abstract: Weight pruning is a common technique for compressing large neural networks. We focus on the challenging post-training one-shot setting, where a pre-trai

Simon Willison discusses results from a 'flamingo on a unicycle' test on his blog, likely comparing AI model performance including Qwen. The post appears to reference previously undisclosed or unconve

arXiv:2604.13853v1 Announce Type: new Abstract: Safe and explainable motion planning remains a central challenge in autonomous driving. While rule-based planners offer predictable and explainable beha

Thunderbolt is a new open-source AI client from Mozilla-owned MZLA Technologies aimed at enterprises who want to run self-hosted chatbots on their own infrastructure. The platform allows organizations

Kyle Orland / Ars Technica: Mozilla launches Thunderbolt, an open-source AI client for users and businesses who want to run their own self-hosted AI infrastructure, available on GitHub — Mozilla is th

arXiv:2505.07591v2 Announce Type: replace Abstract: Instruction following refers to the ability of large language models (LLMs) to generate outputs that satisfy all specified constraints. Existing res

arXiv:2604.13328v1 Announce Type: new Abstract: Pathology reports serve as the definitive record for breast cancer staging, yet their unstructured format impedes large-scale data curation. While Large

New insane model from Jackrong on @huggingface 🤯 Qwen3.5-9B-GLM5.1-Distill-v1 🧠 Distilled on GLM-5.1 reasoning ⚙️ Deeper thinking than base model 🧪 Benchmarks coming soon ✅ Fits on 8GB VRAM ✍️ New mod

Gemini has been updated to generate highly personalized images by accessing a user's Google Photos and personal preferences. This new feature allows users to simply request scenes featuring themselves

Head straight for orbit with GeForce NOW — no space helmet required. PRAGMATA, Capcom’s long-awaited sci-fi action adventure, touches down on GeForce NOW the same day it launches worldwide. The futuri

arXiv:2604.13121v1 Announce Type: new Abstract: Locating and intercepting a moving target from possibly delayed, intermittent sensory signals is a paradigmatic problem in decision-making under uncerta

🎬 Ollama Gemma Day Recap: SGLang at the Ollama Gemma 4 Party in Palo Alto 🍾 Last night, @ollama hosted a packed Gemma Day at the Palo Alto office alongside the @GoogleDeepMind Gemma team. SGLang was i

arXiv:2604.13740v1 Announce Type: new Abstract: We propose a new partial-observability model for online learning problems where the learner, besides its own loss, also observes some noisy feedback abo

Megan Morrone / Axios: OpenAI launches GPT-Rosalind, an AI model for life sciences research, including drug discovery, as a research preview for customers such as Moderna and Amgen — OpenAI announced

OpenAI is beefing up its agentic coding and development system, Codex, with a suite of updates that let it use your computer, generate images, and remember from past experiences. The package of update

arXiv:2603.06885v2 Announce Type: replace Abstract: Trachoma remains the leading infectious cause of blindness worldwide, with Sub-Saharan Africa bearing over 85% of the global burden and Ethiopia alo

arXiv:2604.13672v1 Announce Type: new Abstract: The `spotoptim` package implements surrogate-model-based optimization of expensive black-box functions in Python. Building on two decades of Sequential

Opus 4.7 feels more intelligent, agentic, and precise than 4.6. It took a few days for me to learn how to work with it effectively, to fully take advantage of its new capabilities. Will post a few mor

Opus 4.7 is a model I’ve loved working with in Claude Code. It’s more agentic and instruction following but also incredibly smart and creative. I think it takes a slight adjustment to get used to, but

Opus 4.7 is in Claude Code today. It's more agentic, more precise, and a lot better at long-running work. It carries context across sessions and handles ambiguity much better. Introducing Claude Opus

Opus 4.7 is now supported in Hermes Agent 🚀🚀 Introducing Claude Opus 4.7, our most capable Opus model yet. It handles long-running tasks with more rigor, follows instructions more precisely, and verif

arXiv:2604.13656v1 Announce Type: new Abstract: The statistical essence of the Transformer architecture has long remained elusive: Is it a universal approximator, or a neural network version of known

Our Life Sciences model series is available as a research preview starting today for qualified customers including @Amgen, @moderna_tx, the @AllenInstitute, and @thermofisher Scientific through ChatGP

arXiv:2604.13252v1 Announce Type: new Abstract: Anomaly detection aims to identify observations that deviate from expected behavior. Because anomalous events are inherently sparse, most frameworks are

arXiv:2604.13560v1 Announce Type: new Abstract: Multi-task learning (MTL) improves generalization and data efficiency by jointly learning related tasks through shared representations. In the widely us

arXiv:2604.07662v2 Announce Type: replace-cross Abstract: Monotone variational inequalities (VIs) provide a unifying framework for convex minimization, equilibrium computation, and convex-concave sadd

arXiv:2604.14010v1 Announce Type: cross Abstract: Supervised Fine-Tuning (SFT) of large language models often suffers from task interference and catastrophic forgetting. Recent approaches alleviate th

arXiv:2604.13153v1 Announce Type: new Abstract: 3D Gaussian Splatting (3DGS) has recently enabled highly photorealistic 3D reconstruction from casually captured multi-view images. However, this access

arXiv:2604.13791v1 Announce Type: new Abstract: Accurate lesion segmentation in ultrasound images is essential for preventive screening and clinical diagnosis, yet remains challenging due to low contr

arXiv:2604.13356v1 Announce Type: new Abstract: Mechanisms for continued self-improvement of language models without external supervision remain an open challenge. We propose Peer-Predictive Self-Trai

arXiv:2604.13074v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) serve as daily assistants for millions. However, their ability to generate responses aligned with individual pr

arXiv:2601.03173v2 Announce Type: replace Abstract: Time pressure critically influences risky maneuvers and crash proneness among powered two-wheeler riders, yet its prediction remains underexplored i

Qwen 3.6 is here, and open-source! Run it locally with improved agentic coding capabilities. Try it with Claude Code: ollama launch claude --model qwen3.6 Try it with OpenClaw: ollama launch openclaw

For anyone who has been (inadvisably) taking my pelican riding a bicycle benchmark seriously as a robust way to test models, here are pelicans from this morning's two big model releases - Qwen3.6-35B-

arXiv:2604.05096v2 Announce Type: replace Abstract: Large language models (LLMs) acquire most of their knowledge during pretraining, which ties them to a fixed snapshot of the world and makes adaptati

arXiv:2505.19054v2 Announce Type: replace Abstract: Modern learning-based locomotion controllers typically rely on fully trainable deep neural networks with a large number of parameters. This paper st

arXiv:2604.13213v1 Announce Type: cross Abstract: Rare events such as conformational changes in biomolecules, phase transitions, and chemical reactions are central to the behavior of many physical sys

arXiv:2604.13730v1 Announce Type: new Abstract: Continual learning enables models to acquire new knowledge over time while retaining previously learned capabilities. However, its application to text-t

arXiv:2604.13064v1 Announce Type: new Abstract: Skill ecosystems have emerged as an increasingly important layer in Large Language Model (LLM) agent systems, enabling reusable task packaging, public d

Replit Agent 4 has been upgraded to use Claude Opus 4.7, an advanced AI model, enhancing its code generation and problem-solving capabilities. The company is offering a 50% discount for a limited time

Response. Hey Ethan! Sean here, PM on http://Claude.ai - thanks for the feedback. This isn't a router, this is the model being trained to decide when to think based on the context -- we've been runnin

arXiv:2604.13993v1 Announce Type: cross Abstract: Physical reasoning over visual inputs demands tight integration of visual perception, domain knowledge, and multi-step symbolic inference. Yet even st

arXiv:2604.13602v1 Announce Type: new Abstract: Reinforcement Learning from Human Feedback (RLHF) and related alignment paradigms have become central to steering large language models (LLMs) and multi

arXiv:2604.13531v1 Announce Type: cross Abstract: Graphical User Interface (GUI) agents show strong capabilities for automating web tasks, but existing interactive benchmarks primarily target benign,

arXiv:2603.22126v3 Announce Type: replace Abstract: Deploying learned robot manipulation policies in industrial settings requires rigorous pre-deployment validation, yet exhaustive testing across high

arXiv:2604.13525v1 Announce Type: cross Abstract: The robust low-rank tensor completion problem addresses the challenge of recovering corrupted high-dimensional tensor data with missing entries, outli