model releases

WorldFly: A World-Model-Based Vision-Language-Action Model for UAV Navigation

model-releasesarxiv-cs-ai

arXiv:2606.06147v1 Announce Type: new Abstract: End-to-end Vision-Language-Action (VLA) models have shown promise in UAV navigation. However, existing approaches typically rely on historical observati

6 Jun 2026

Model Releases

You can’t let this happen, @DavidSacks, cc @elonmusk. It’s WWWIII but with AI. Nobody wins.

DGX agent

You can’t let this happen, @DavidSacks, cc @elonmusk. It’s WWWIII but with AI. Nobody wins. ⚠️⚠️ Seismic shift ⚠️⚠️ It’s a good day to be Mistral. Nobody is going to trust an American AI company that

model-releasesgary-marcus--x

6 Jun 2026

Your margin is my opportunity: AI version… The biggest surprise of 2026 is that the capability gap between the best open-weight/source model…

model-releasesclem-delangue--x

Your margin is my opportunity: AI version… The biggest surprise of 2026 is that the capability gap between the best open-weight/source models and the best closed models has narrowed much faster than t

6 Jun 2026

29 more Starlink satellites. Over 10,000 in orbit now.

model-releaseselon-musk--x

SpaceX launched 29 additional Starlink satellites, bringing the total constellation to over 10,000 satellites in orbit. This milestone represents a significant expansion of the Starlink mega-constella

3D Underwater Path Planning via Generative Flow Field Surrogates

arXiv:2606.06077v1 Announce Type: new Abstract: Autonomous underwater vehicle (AUV) launch and recovery (LAR) into the hull of an advancing host platform requires traversal of a complex, three-dimensi

A Novel Method with Encoder-Decoder for Cross-Sensor Adaptation in Surface Shape Sensing with Sparse Strain Sensors

arXiv:2606.05903v1 Announce Type: new Abstract: Performance variations in sensor arrays, caused by intrinsic differences or installation conditions, can lead to inconsistent results during shape sensi

A research team that includes Huawei says it successfully used Huawei's Ascend 910C chips for DeepSeek V4 Pro model's post-training, amid increased US sanctions (Coco Feng/South China Morning Post)

Coco Feng / South China Morning Post: A research team that includes Huawei says it successfully used Huawei's Ascend 910C chips for DeepSeek V4 Pro model's post-training, amid increased US sanctions —

model-releasestechmeme

AdaPlanBench: Evaluating Adaptive Planning in Large Language Model Agents under World and User Constraints

arXiv:2606.05622v1 Announce Type: new Abstract: Planning for real-world problems by language models often involves both world and user constraints, which may not be fully specified upfront and are pro

Adaptive Tokenisation Via Temporal Redundancy Masking And Latent Inpainting

arXiv:2606.06158v1 Announce Type: new Abstract: Adaptive video tokenisation seeks to dynamically allocate token budgets based on the underlying visual complexity of a sequence. Current continuous-regi

Agents' Last Exam

arXiv:2606.05405v1 Announce Type: cross Abstract: Recent AI systems have achieved strong results on a wide range of benchmarks, yet these gains have not translated into economically meaningful deploym

// Agents' Last Exam // Agents' Last Exam is a living benchmark of over 1,000 economically valuable tasks, built with 250+ industry experts …

// Agents' Last Exam // Agents' Last Exam is a living benchmark of over 1,000 economically valuable tasks, built with 250+ industry experts and mapped to the U.S. federal occupational taxonomy. The ha

model-releasesdair-ai--x

Aligning Tree-Search Policies with Fixed Token Budgets in Test-Time Scaling of LLMs

arXiv:2602.09574v2 Announce Type: replace Abstract: Tree-search decoding is an effective form of test-time scaling for large language models (LLMs), but real-world deployment often imposes a fixed per

Almieyar-Oryx-BloomBench: A Bilingual Multimodal Benchmark for Cognitively Informed Evaluation of Vision-Language Models

arXiv:2606.05531v1 Announce Type: cross Abstract: Despite the rapid progress of Vision-Language Models (VLMs), the field lacks benchmarks that rigorously diagnose their true reasoning abilities and ch

An Embarrassingly Simple Detector for Model Extraction Attacks in Large Language Model API Traffic

arXiv:2606.05725v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly deployed through hosted APIs, making model extraction a practical threat to model ownership and service

An issue caused some user accounts to be incorrectly suspended. We’re restoring access and working through related subscription and credit i…

OpenAI experienced a technical issue that resulted in some user accounts being incorrectly suspended. The company announced it is working to restore access to affected accounts and address related iss

model-releasesopenai--x

ArcANE: Do Role-Playing Language Agents Stay in Character at the Right Time?

arXiv:2606.05553v1 Announce Type: new Abstract: Role-playing language agents (RPLAs) should play characters whose values and behavior evolve as the story progresses, not maintain a fixed persona. Exis

Arena AI Agentic User Benchmark Ranking

Arena AI's agentic benchmark ranks AI models on how well they orchestrate tools for real-world agentic tasks, based on signals like tool reliability, task completion, and steerability. The leaderboard

model-releasesr-chatgpt

Asuka-Bench: Benchmarking Code Agents on Underspecified User Intent and Multi-Round Refinement

arXiv:2606.05920v1 Announce Type: cross Abstract: Existing code-generation benchmarks score a single mapping from a complete prompt to a one-shot output. However, real web development is different. Us

At least until (if?) rapid improvement stops, it seems less likely someone is going to catch the Big Three AI Labs. Microsoft and Meta relea…

model-releasesethan-mollick--x

At least until (if?) rapid improvement stops, it seems less likely someone is going to catch the Big Three AI Labs. Microsoft and Meta released their models, which were fine, but not frontier. SpaceX

Augment Code launches Cosmos to bring agentic AI software development to teams

model-releasessiliconangle

Augment Code Computing Inc., an artificial intelligence agent platform provider, Thursday announced the launch of Cosmos, a service it says is designed to push beyond the era of individual AI coding a

AURA: Intent-Directed Probing for Implicit-Need Surfacing in Situated LLM Agents

arXiv:2606.05557v1 Announce Type: new Abstract: A situated query like 'where is Lin Wei?' often encodes more than its literal content: the user may also want to know whether Lin Wei is free, in a good

Before the week ends, let's acknowledge one of the most INSANE week ever for open AI, with 25+ notable open-weight drops across every modali…

model-releasesclem-delangue--x

Before the week ends, let's acknowledge one of the most INSANE week ever for open AI, with 25+ notable open-weight drops across every modality: 🧠 LLMs → NVIDIA Nemotron 3 Ultra: 550B hybrid Mamba-MoE,

Benchmarking Open-Source Layout Detection Models for Data Snapshot Extraction from Institutional Documents

arXiv:2606.06242v1 Announce Type: new Abstract: Institutional documents contain substantial amounts of operational and analytical information embedded within figures and tables. Current approaches for

Better Literary Translation: A Multi-Aspect Data Generation and LLM Training Approach

arXiv:2606.05924v1 Announce Type: new Abstract: Literary translation poses unique challenges due to the scarcity of high-quality annotated data and the need to balance expression fluency with literary

Biomazon: A Multimodal Dataset for 3D Forest Structure and Biomass Modeling in the Amazon Basin

arXiv:2606.05368v1 Announce Type: new Abstract: Accurate, spatially explicit characterization of tropical forest structure is essential for carbon accounting and ecosystem monitoring, yet most ML pipe

Breaking Time: A Fully Gaussian Framework for Distributed and Continuous-Time SLAM

arXiv:2606.06250v1 Announce Type: new Abstract: Continuous-time SLAM provides a principled framework for fusing heterogeneous sensors while estimating smooth trajectories, and is particularly well-sui

CamFlow+: Hybrid Motion Bases for 2D Camera Motion Estimation with Stabilization Applications

arXiv:2606.05915v1 Announce Type: new Abstract: Estimating 2D camera motion is fundamental to computer vision and computational photography. Existing homography-based methods work well for planar scen

Can LLMs Be Constrained to the Past? Improving Knowledge Cutoff through Recall-Based Prompting

arXiv:2606.05804v1 Announce Type: new Abstract: Prompted knowledge cutoff instructs a large language model (LLM) to act as if information beyond a specified cutoff date were unavailable. However, prio

CHALIS: A Challenge Dataset for Language Identification in Difficult Scenarios

arXiv:2606.06088v1 Announce Type: new Abstract: We present CHALIS (Challenging Language Identification Samples), a new benchmark dataset explicitly designed to address difficult cases in language iden

Channel-Wise Mixed-Precision Quantization for Large Language Models

arXiv:2410.13056v4 Announce Type: replace Abstract: Large Language Models (LLMs) have demonstrated remarkable success across a wide range of language tasks, but their deployment on edge devices remain

CLASH: Evaluating Language Models on Judging High-Stakes Dilemmas from Multiple Perspectives

arXiv:2504.10823v4 Announce Type: replace Abstract: Navigating dilemmas involving conflicting values is challenging even for humans in high-stakes domains, let alone for AI, yet prior work has been li

CLEAR: Cognition and Latent Evaluation for Adaptive Routing in End-to-End Autonomous Driving

arXiv:2606.06219v1 Announce Type: new Abstract: End-to-end autonomous driving models often struggle to balance multi-modal maneuver generation with real-time inference constraints. While diffusion mod

CLFEC: A New Task for Unified Linguistic and Factual Error Correction in paragraph-level Chinese Professional Writing

arXiv:2602.23845v2 Announce Type: replace Abstract: Chinese text correction has traditionally focused on spelling and grammar, while factual error correction is usually treated separately. However, in

Code2LoRA: Hypernetwork-Generated Adapters for Code Language Models under Software Evolution

arXiv:2606.06492v1 Announce Type: cross Abstract: Code language models need repository-level context to resolve imports, APIs, and project conventions. Existing methods inject this knowledge as long i

Coding with 'Enemy': Can Human Developers Detect AI Agent Sabotage?

arXiv:2606.05647v1 Announce Type: cross Abstract: AI coding agents are increasingly embedded in real-world software development, collaborating with human developers while gaining broader access to cod

CollabBench: Benchmarking and Unleashing Collaborative Ability of LLMs with Diverse Players via Proactive Engagement

arXiv:2606.05793v1 Announce Type: new Abstract: While LLM-based agents excel at individual tasks, effective collaboration with realistic human partners remains challenging. Most of the existing conver

CoMoL: Efficient Mixture of LoRA Experts via Dynamic Core Space Merging

arXiv:2603.00573v2 Announce Type: replace Abstract: Large language models (LLMs) achieve remarkable performance on diverse downstream and domain-specific tasks via parameter-efficient fine-tuning (PEF

Compress-Distill: Reasoning Trace Compression for Efficient Knowledge Distillation

arXiv:2606.05988v1 Announce Type: cross Abstract: Reasoning models produce long chain-of-thought traces that are costly to distill and encourage verbose student outputs. We study post-hoc compression

Contextualized Prompting For Stance Detection On Social Media

arXiv:2606.06022v1 Announce Type: new Abstract: Stance detection on social media is challenging due to short, noisy, and context-dependent language. While large language models (LLMs) show zero-shot g

Continual Learning Bench: Evaluating Frontier AI Systems in Real-World Stateful Environments

arXiv:2606.05661v1 Announce Type: cross Abstract: Continual learning, the ability of AI systems to improve through sequential experience, has attracted substantial interest, but no high-quality benchm

Critical context on the new Anthropic blog: 1, AGI is harder than RSI (as used below). AGI: machine can do anything human can do, autonomo…

model-releasesgary-marcus--x

Critical context on the new Anthropic blog: 1, AGI is *harder* than RSI (as used below). AGI: machine can do anything human can do, autonomously [not achieved] RSI (as used below): AI is a useful codi

Dense Contexts Are Hard Contexts: Lexical Density Limits Effective Context in LLMs

arXiv:2606.06203v1 Announce Type: new Abstract: Input length and the position of relevant information are widely cited as the primary causes of degraded LLM long-context performance. Here, we study le

DisasterBench: A Multimodal Benchmark for UAV-Based Disaster Response in Complex Environments

arXiv:2606.06217v1 Announce Type: new Abstract: When a disaster unfolds, responders must answer not only what is happening, but also why it is happening, what will happen next, and what to do now, oft

Do MLLMs Capture How Interfaces Guide User Behavior? A Benchmark for Multimodal UI/UX Design Understanding

arXiv:2505.05026v5 Announce Type: replace Abstract: User interface (UI) design goes beyond visuals to shape user experience (UX), underscoring the shift toward UI/UX as a unified concept. While recent

DocHop-QA: Towards Multi-Hop Reasoning over Multimodal Document Collections

arXiv:2508.15851v2 Announce Type: replace Abstract: Despite rapid progress in large language models (LLMs), current QA benchmarks still overlook the core challenge of real-world scientific information

Domain-Aware Mispronunciation Detection and Diagnosis Using Language-Specific Statistical Graphs

arXiv:2606.05569v1 Announce Type: new Abstract: Mispronunciation Detection and Diagnosis (MDD) has gained increasing importance in computer-assisted language learning and speech technology in recent y

Domain-Conditioned Safety in Frontier Computer-Using Agents: A 793-Episode Browser Benchmark, a Coding-Domain Cross-Reference, and a Reproducibility Audit of Recent Red-Teaming

arXiv:2606.05233v1 Announce Type: cross Abstract: Recent computer-using-agent (CUA) red-teaming papers report prompt-injection attack success rates (ASR) of 42-98%, but these headline numbers cluster

Drive-KD: Multi-Teacher Distillation for VLMs in Autonomous Driving

arXiv:2601.21288v2 Announce Type: replace-cross Abstract: Autonomous driving is an important and safety-critical task, and recent advances in LLMs/VLMs have opened new possibilities for reasoning and