model releases

Two is better than one: A Collapse-free Multi-Reward RLIF Training Framework

arXiv:2605.22620v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has substantially improved the reasoning ability of LLMs, but often depends on external supervis

Model Releases

Ultra-High-Definition Image Quality Assessment via Graph Representation Learning

DGX agent

arXiv:2605.22192v1 Announce Type: new Abstract: Blind image quality assessment (BIQA) for ultrahighdefinition (UHD) images remains challenging because native-resolution inference is computationally ex

model-releasesarxiv-cs-cv

22 May 2026

Understanding Data Temporality Impact on Large Language Models Pre-training

arXiv:2605.22769v1 Announce Type: new Abstract: Large language models (LLMs) are typically trained on shuffled corpora, yielding models whose knowledge is frozen at train time and whose temporal groun

UniVL: Unified Vision-Language Embedding for Spatially Grounded Contextual Image Generation

arXiv:2605.21611v1 Announce Type: new Abstract: We introduce spatially grounded contextual image generation, a controllable image generation task that reframes the conditioning paradigm. Instead of su

VDE Bench: Evaluating The Capability of Image Editing Models to Modify Visual Documents

arXiv:2602.00122v2 Announce Type: replace Abstract: In recent years, image editing models have made significant progress, enabling users to manipulate visual content in a flexible and interactive mann

VEELA: A Clinically-Constrained Benchmark for Liver Vessel Segmentation in Computed Tomography Angiography

arXiv:2605.22357v1 Announce Type: new Abstract: Accurate segmentation of hepatic and portal vessels in contrast-enhanced computed tomography angiography (CTA) remains challenging due to complex vascul

VGenST-Bench: A Benchmark for Spatio-Temporal Reasoning via Active Video Synthesis

arXiv:2605.22570v1 Announce Type: new Abstract: Spatio-temporal reasoning is a core capability for Multimodal Large Language Models (MLLMs) operating in the real world. As such, evaluating it precisel

VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction

arXiv:2602.13294v3 Announce Type: replace Abstract: Evaluating whether Multimodal Large Language Models (MLLMs) genuinely reason about physical dynamics remains challenging. Most existing benchmarks r

Visual-Advantage On-Policy Distillation for Vision-Language Models

arXiv:2605.21924v1 Announce Type: new Abstract: On-policy knowledge distillation has proven effective for language models, yet its application to vision-language models (VLMs) remains underexplored. W

We are making our discount permanent! 🎉 Enjoy building with DeepSeek-V4-Pro and bring your innovative ideas to life! 🚀

model-releasesjeremy-howard--x

DeepSeek has announced a permanent discount for its DeepSeek-V4-Pro model, encouraging developers to build and innovate with the platform. The announcement was made via social media and emphasizes the

We desperately need better ways of evaluating models. Something that shows how helpful they are at working hand-in-hand with humans to help …

model-releasesjeremy-howard--x

We desperately need better ways of evaluating models. Something that shows how helpful they are at working hand-in-hand with humans to help them get stuff done in a cooperative/iterative way. The Clau

We’re taking suggestions on what you want to see next week ✍️

OpenAI solicited community feedback on X regarding content or features they should prioritize in the following week. This post reflects OpenAI's practice of engaging their audience to guide product de

model-releasesopenai--x

When Cases Get Rare: A Retrieval Benchmark for Off-Guideline Clinical Question Answering

arXiv:2605.21807v1 Announce Type: new Abstract: Across medical specialties, clinical practice is anchored in evidence-based guidelines that codify best studied diagnostic and treatment pathways. These

Wordle 1,797 4/6 ⬛⬛🟨🟨🟩 ⬛⬛⬛🟨⬛ 🟨🟨🟨⬛🟩 🟩🟩🟩🟩🟩

model-releasesanthropic--x

This post documents a Wordle game result (puzzle #1,797) where the player achieved a solution in 4 attempts, using color-coded emoji feedback (⬛ for incorrect letters, 🟨 for correct letters in wrong p

X-Token: Projection-Guided Cross-Tokenizer Knowledge Distillation

arXiv:2605.21699v1 Announce Type: cross Abstract: Cross-tokenizer knowledge distillation allows a student model to learn from teachers with incompatible vocabularies. Prior work operates on hidden sta

Yesterday we released Aleph 2.0, our upgraded video editing model that lets you change exactly what you want while keeping everything else t…

model-releasescristobal-valenzuela--x

Yesterday we released Aleph 2.0, our upgraded video editing model that lets you change exactly what you want while keeping everything else the same. Available inside our new Edit Studio, you can work

A Deployment Audit of Release-Side Risk in Conformal Triage under Prevalence Shift

arXiv:2605.20956v1 Announce Type: new Abstract: Conformal triage converts predictive scores into deployment actions that either release a case, flag it for urgent attention, or defer it to human revie

A Free Lunch in LLM Compression: Revisiting Retraining after Pruning

arXiv:2510.14444v3 Announce Type: replace Abstract: Post-training pruning can substantially reduce LLM inference costs, but it often degrades quality unless the remaining weights are adapted. Since gl

A strongly annotated passive acoustic dataset for tropical bird monitoring

arXiv:2605.20578v1 Announce Type: cross Abstract: Passive acoustic monitoring enables continuous, non-invasive biodiversity assessment across diverse ecosystems. The scale of these datasets has driven

A Unified Framework for Uncertainty-Aware Explainable Artificial Intelligence: A Case Study in Power Quality Disturbance Classification

arXiv:2605.21114v1 Announce Type: new Abstract: Post-hoc explainable AI (XAI) methods typically produce deterministic attribution maps, whereas Bayesian neural networks (BNNs) induce a distribution ov

ACL-Verbatim: hallucination-free question answering for research

arXiv:2605.21102v1 Announce Type: new Abstract: Academic researchers need efficient and reliable methods for collecting high-quality information from trusted sources, but modern tools for AI-assisted

Ada2MS: A Hybrid Optimization Algorithm Based on Exponential Mixing of Elementwise and Global Second-Moment Estimates

arXiv:2605.20533v1 Announce Type: new Abstract: Optimization algorithms are core methods by which machine learning models iteratively minimize loss functions, update parameters, learn from data, and i

Adobe, Canva, and CapCut announce Gemini integrations to let users access the companies' image and video editing tools within the Gemini app (James Peckham/PCMag)

James Peckham / PCMag: Adobe, Canva, and CapCut announce Gemini integrations to let users access the companies' image and video editing tools within the Gemini app — Adobe, Canva, and CapCut all plan

model-releasestechmeme

Adversarial Robustness in One-Stage Learning-to-Defer

arXiv:2510.10988v2 Announce Type: replace-cross Abstract: Learning-to-Defer (L2D) enables hybrid decision-making by routing inputs either to a predictor or to external experts. While promising, L2D is

AgentAtlas: Beyond Outcome Leaderboards for LLM Agents

arXiv:2605.20530v1 Announce Type: cross Abstract: Large language model agents now act on codebases, browsers, operating systems, calendars, files, and tool ecosystems, but the benchmarks used to evalu

Agentic Physical AI toward a Domain-Specific Foundation Model for Nuclear Reactor Control

arXiv:2512.23292v3 Announce Type: replace-cross Abstract: The prevailing paradigm in AI for physical systems (scaling general-purpose foundation models toward universal multimodal reasoning) confronts

AGPO: Adaptive Group Policy Optimization with Dual Statistical Feedback

arXiv:2605.20722v1 Announce Type: new Abstract: Reinforcement learning improves LLM reasoning, but PPO/GRPO typically use fixed clipping and decoding temperature, which makes training brittle and tuni

AI-Assisted Scientific Assessment: A Case Study on Climate Change

arXiv:2602.09723v2 Announce Type: replace Abstract: The emerging paradigm of AI co-scientists focuses on tasks characterized by repeatable verification, where agents explore search spaces in 'guess an

AMAR: Lightweight Attention-Based Multi-User Activity Recognition from Wi-Fi CSI

arXiv:2605.20649v1 Announce Type: cross Abstract: Wi-Fi-based human activity recognition (HAR) has emerged as a promising approach for contactless sensing, leveraging channel state information (CSI) c

An exponential mechanism based on quadratic approximations for fine-tuning machine learning models with privacy guarantees

arXiv:2605.20521v1 Announce Type: new Abstract: Fine-tuning adapts a pretrained machine learning model to a small, sensitive dataset, but this process risks memorizing individual new data points, maki

Anatomy of Agentic Memory: Taxonomy and Empirical Analysis of Evaluation and System Limitations

arXiv:2602.19320v2 Announce Type: replace Abstract: Agentic memory systems enable large language model (LLM) agents to maintain state across long interactions, supporting long-horizon reasoning and pe

AnimeAdapter: Fine-grained and Consistent Zero-shot Anime Character Generation

arXiv:2605.20237v1 Announce Type: new Abstract: We present a lightweight appearance adapter for Stable Diffusion that enables controllable and consistent anime character generation under diverse editi

Another win for open-source robotics! 🔥 @huggingface just released a fully open-source humanoid robot, and you can build one for $2,500. I'…

model-releasesclem-delangue--x

Another win for open-source robotics! 🔥 @huggingface just released a fully open-source humanoid robot, and you can build one for $2,500. I'm a huge advocate of open-source in robotics space. Why? Robo

Anthropic’s Code with Claude showed off coding’s future—whether you like it or not

model-releasesmit-tech-review

The vibes were strong at Code with Claude, Anthropic’s two-day event for software developers in London that kicked off on May 19, the same day as Google’s I/O in Palo Alto. (A coincidence, not a flex,

APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents

arXiv:2605.21240v1 Announce Type: new Abstract: LLM agents have shown strong performance across a wide range of complex tasks, including interactive environments that require long-horizon decision mak

API Keys Are Open Secrets

model-releasesgoogle-cloud-ai

Today, AI services rely heavily on API keys. To run AI agents, users provide API keys that signify paid tokens, subscriptions, or paid accounts. While API keys are easy to use, it is just as easy to u

APM: Evaluating Style Personalization in LLMs with Arbitrary Preference Mappings

arXiv:2605.21063v1 Announce Type: new Abstract: Typical LLM responses tend to follow a default style, even though users often have distinct preferences regarding tone, verbosity, and formality that th

Approximation Theory for Neural Networks: Old and New

arXiv:2605.21451v1 Announce Type: new Abstract: Universal approximation theorems provide a mathematical explanation for the expressive power of neural networks. They assert that, under mild conditions

ArchSIBench: Benchmarking the Architectural Spatial Intelligence of Vision-Language Models

arXiv:2605.20837v1 Announce Type: new Abstract: Architectural spatial intelligence, the ability to recognize and infer architectural space, is fundamental to tasks such as robot navigation, embodied i

AttriStory: Fine-grained Attribute Realization for Visual Storytelling with Diffusion Models

arXiv:2605.20777v1 Announce Type: new Abstract: Visual storytelling with diffusion models has made impressive strides in maintaining character consistency across narrative scenes. However, a critical

Automated ICD Classification of Psychiatric Diagnoses: From Classical NLP to Large Language Models

arXiv:2605.21154v1 Announce Type: new Abstract: Mental health has become a global priority, leading to a massive administrative burden in the coding of clinical diagnoses. This study proposes the auto

Batched Single-Index Global Multi-Armed Bandits with Covariates

arXiv:2503.00565v3 Announce Type: replace-cross Abstract: The multi-armed bandits (MAB) framework is a widely used approach for sequential decision-making, where a decision-maker selects an arm in eac

Benchmarking Empirical and Learning-Based Approaches for Feedforward Steering Control in Autonomous Racing

model-releasesarxiv-cs-ro

arXiv:2605.21111v1 Announce Type: new Abstract: Feedforward steering control is a key component of hierarchical control architectures for autonomous racing. The goal is to reduce steering corrections

Break the context window barrier with Amazon Bedrock AgentCore

model-releasesaws-ml-blog

In this post, you will learn how to implement Recursive Language Models (RLM) using Amazon Bedrock AgentCore Code Interpreter and the Strands Agents SDK. By the end, you will know how to process docum

Bridging Structure and Language: Graph-Based Visual Reasoning for Autonomous Road Understanding

arXiv:2605.20942v1 Announce Type: new Abstract: Structured road understanding of lane geometry, topology, and traffic element relationships is foundational to safe autonomous driving. While vision-lan

Bugcrowd launches reinforcement learning environments to train AI on real software vulnerabilities

model-releasessiliconangle

Crowdsourced cybersecurity company Bugcrowd Inc. today launched Reinforcement Learning Environments, a new offering that lets frontier artificial intelligence labs train models on real vulnerable soft

Build a Coding Assistant with Weaviate MCP: RAG over Code & Docs

This guide demonstrates how to build a coding assistant using Weaviate's Model Context Protocol (MCP) integration, enabling retrieval-augmented generation (RAG) capabilities over codebases and documen

model-releasesweaviate

Build AI agents for business intelligence with Amazon Bedrock AgentCore

model-releasesaws-ml-blog

In this post, we show you how OPLOG developed three AI agents using the Strands Agents SDK, deployed them to Amazon Bedrock AgentCore, and integrated Amazon Bedrock with Anthropic’s Claude Sonnet and