Search: “models”

Frontier Coding Agents Can Now Implement an AlphaZero Self-Play Machine Learning Pipeline For Connect Four That Performs Comparably to an External Solver

arXiv:2604.25067v1 Announce Type: cross Abstract: Forecasting when AI systems will become capable of meaningfully accelerating AI research is a central challenge for AI safety. Existing benchmarks mea

Model Releases

GAIA-v2-LILT: Multilingual Adaptation of Agent Benchmark beyond Translation

DGX agent

arXiv:2604.24929v1 Announce Type: new Abstract: Agent benchmarks remain largely English-centric, while their multilingual versions are often built with machine translation (MT) and limited post-editin

model-releasesarxiv-cs-cl

29 Apr 2026

Gemini now can create documents, and it is a nice start, but not up to the frontier yet, as you can see from my 'LBO of Hogwarts' test. Powe…

model-releasesethan-mollick--x

Gemini now can create documents, and it is a nice start, but not up to the frontier yet, as you can see from my 'LBO of Hogwarts' test. PowerPoints are substantially worse than NotebookLM, spreadsheet

Google Photos launches an AI try-on feature for clothes you already have

model-releasesthe-verge-ai

Google Photos is launching a new AI-powered feature you can use to virtually try on clothes you already have. Using the photos in your gallery, Google will create a virtual 'wardrobe,' allowing you to

Google says paid subscriptions reached 350M in Q1, up 25M QoQ, driven by YouTube and Google One, while Gemini Enterprise paid MAUs grew 40% QoQ (Sarah Perez/TechCrunch)

Sarah Perez / TechCrunch: Google says paid subscriptions reached 350M in Q1, up 25M QoQ, driven by YouTube and Google One, while Gemini Enterprise paid MAUs grew 40% QoQ — Google has added another 25

model-releasestechmeme

Gradient-Direction Sensitivity Reveals Linear-Centroid Coupling Hidden by Optimizer Trajectories

arXiv:2604.25143v1 Announce Type: new Abstract: We show that replacing the rolling SVD of AdamW updates with a rolling SVD of loss gradients changes the diagnostic by 1-2 orders of magnitude. Performi

Tools

Granite 4.1 LLMs: How They’re Built

IBM's Granite 4.1 is a series of large language models with improved capabilities over previous versions, likely trained on diverse datasets and optimized for various enterprise and general-purpose ta

toolshugging-face

GraphPL: Leveraging GNN for Efficient and Robust Modalities Imputation in Patchwork Learning

arXiv:2604.25352v1 Announce Type: new Abstract: Current research on distributed multi-modal learning typically assumes that clients can access complete information across all modalities, which may not

Hierarchical Reinforcement Learning for the Dynamic VNE with Alternatives Problem

arXiv:2512.05207v2 Announce Type: replace-cross Abstract: Virtual Network Embedding (VNE) is a key enabler of network slicing, yet most formulations assume that each Virtual Network Request (VNR) has

Tutorials

How Can Reinforcement Learning Achieve Expert-level Placement?

arXiv:2604.25191v1 Announce Type: cross Abstract: Chip placement is a critical step in physical design. While reinforcement learning (RL)-based methods have recently emerged, their training primarily

tutorialsarxiv-cs-lg

i love that the team does stuff like this

model-releasessam-altman--x

i love that the team does stuff like this Codex is not like claude code. if you know the limit is going to end, like last 10 to 8%, give an very long run task, and even after the limit got ever, it wi

Image Classification via Random Dilated Convolution with Multi-Branch Feature Extraction and Context Excitation

arXiv:2604.25188v1 Announce Type: new Abstract: Image classification remains a fundamental yet challenging task in computer vision, particularly when fine-grained feature extraction and background noi

Introducing remote agents in Vibe and Mistral Medium 3.5. You can now launch remote agents in the cloud, including from the CLI or Le Chat. …

model-releasesmistral-ai--x

Introducing remote agents in Vibe and Mistral Medium 3.5. You can now launch remote agents in the cloud, including from the CLI or Le Chat. Plus, new Work mode in Le Chat for complex, multi-step tasks

Applications

JaGuard: Position Error Correction of GNSS Jamming with Deep Temporal Graphs

arXiv:2509.14000v4 Announce Type: replace Abstract: Global Navigation Satellite Systems (GNSS) face growing disruption from intentional jamming, undermining critical infrastructure where precise posit

applicationsarxiv-cs-lg

Last week we announced DeepSeek-V4. Today we’re sharing a closer look at DeepSeek-V4 Pro on Together AI: 512K context, controllable reasonin…

model-releasestogether-ai--x

Last week we announced DeepSeek-V4. Today we’re sharing a closer look at DeepSeek-V4 Pro on Together AI: 512K context, controllable reasoning modes, and cached-input pricing for long-context workloads

Safety

Learning from Noisy Preferences: A Semi-Supervised Learning Approach to Direct Preference Optimization

arXiv:2604.24952v1 Announce Type: new Abstract: Human visual preferences are inherently multi-dimensional, encompassing aesthetics, detail fidelity, and semantic alignment. However, existing datasets

safetyarxiv-cs-cv

LongSumEval: Question-Answering Based Evaluation and Feedback-Driven Refinement for Long Document Summarization

arXiv:2604.25130v1 Announce Type: new Abstract: Evaluating long document summaries remains the primary bottleneck in summarization research. Existing metrics correlate weakly with human judgments and

Lookout launches mobile-native tool to expose shadow AI on enterprise devices

model-releasessiliconangle

Cybersecurity company Lookout Inc. today announced the launch of Lookout AI Visibility & Governance, a new mobile-native solution designed to provide organizations with the visibility needed to discov

Metric, inertially aligned monocular state estimation via kinetodynamic priors

arXiv:2511.20496v3 Announce Type: replace Abstract: Accurate state estimation for flexible robotic systems poses significant challenges, particularly for platforms with dynamically deforming structure

researcharxiv-cs-ro

minAction.net: Energy-First Neural Architecture Design -- From Biological Principles to Systematic Validation

arXiv:2604.24805v1 Announce Type: new Abstract: Modern machine learning optimizes for accuracy without explicitly accounting for internal computational cost, even though physical and biological system

Minimax Generalized Cross-Entropy

arXiv:2603.19874v3 Announce Type: replace-cross Abstract: Loss functions play a central role in supervised classification. Cross-entropy (CE) is widely used, whereas the mean absolute error (MAE) loss

Safety

MolReFlect: Towards In-Context Fine-grained Alignments between Molecules and Texts

arXiv:2411.14721v2 Announce Type: replace Abstract: Molecule discovery is a pivotal research field, impacting everything from medicine to materials. Recently, Large Language Models (LLMs) have been wi

safetyarxiv-cs-cl

Industry

Motorola reveals 2026 Razr lineup with modest upgrades and higher prices

Motorola announced its 2026 Razr lineup featuring familiar designs with price increases across all models. The lineup received battery upgrades, with the base Razr gaining a 4,800mAh cell, the Razr+ r

industryars-technica

Multimodal Contextualized Support for Enhancing Video Retrieval System

arXiv:2412.07584v2 Announce Type: replace Abstract: Current video retrieval systems, especially those used in competitions, primarily focus on querying individual keyframes or images rather than encod

researcharxiv-cs-cv

Applications

Near-Optimal Sample Complexities of Divergence-based S-rectangular Distributionally Robust Reinforcement Learning

arXiv:2505.12202v3 Announce Type: replace Abstract: Distributionally robust reinforcement learning (DR-RL) has recently gained significant attention as a principled approach that addresses discrepanci

applicationsarxiv-cs-lg

Pentagon's Digital and AI Chief Cameron Stanley confirms the DoD is expanding its Google Gemini use, saying 'overreliance on one vendor is never a good thing' (Seema Mody/CNBC)

Seema Mody / CNBC: Pentagon's Digital and AI Chief Cameron Stanley confirms the DoD is expanding its Google Gemini use, saying “overreliance on one vendor is never a good thing” — Pentagon AI chief Ca

model-releasestechmeme

Praxy Voice: Voice-Prompt Recovery + BUPS for Commercial-Class Indic TTS from a Frozen Non-Indic Base at Zero Commercial-Training-Data Cost

arXiv:2604.25441v1 Announce Type: cross Abstract: Commercial TTS systems produce near-native Indic audio, but the best open-source bases (Chatterbox, Indic Parler-TTS, IndicF5) trail them on measured

PSP: An Interpretable Per-Dimension Accent Benchmark for Indic Text-to-Speech

arXiv:2604.25476v1 Announce Type: cross Abstract: Standard text-to-speech (TTS) evaluation measures intelligibility (WER, CER) and overall naturalness (MOS, UTMOS) but does not quantify accent. A synt

QB-LIF: Learnable-Scale Quantized Burst Neurons for Efficient SNNs

arXiv:2604.25688v1 Announce Type: new Abstract: Binary spike coding enables sparse and event-driven computation in spiking neural networks (SNNs), yet its 1-bit-per-timestep representation fundamental

Query-Efficient Quantum Approximate Optimization via Graph-Conditioned Trust Regions

arXiv:2604.24803v1 Announce Type: new Abstract: In low-depth implementations of the Quantum Approximate Optimization Algorithm (QAOA), the dominant cost is often the number of objective evaluations ra

Qwen 3.5 from @Alibaba_Qwen is now available on @FireworksAI_HQ Training Platform across the Managed and Training API workflows. Try SFT, DP…

model-releasesfireworks-ai--x

Qwen 3.5 from @Alibaba_Qwen is now available on @FireworksAI_HQ Training Platform across the Managed and Training API workflows. Try SFT, DPO, RL with smart defaults or your own custom loss function w

Local Ai

Qwen3.6 27B on dual RTX 5060 Ti 16GB with vLLM: ~60 tok/s, 204k context working

Qwen3.6 27B is a 27-billion parameter language model that can achieve approximately 60 tokens per second throughput when running on dual RTX 5060 Ti GPUs with 16GB memory each, using the vLLM inferenc

local-air-localllama

Recreated the app with ml intern to compare. Went faster than in cowork, some things are better (ex direct integration in the desktop reachy…

model-releasesclem-delangue--x

Recreated the app with ml intern to compare. Went faster than in cowork, some things are better (ex direct integration in the desktop reachy mini app), some things are worse (had to debut the install)

Representation Paradigms in AI-based 3D Radiological Image Reconstruction: A Systematic Review

arXiv:2504.11349v3 Announce Type: replace Abstract: The demand for high-quality medical imaging in clinical practice and assisted diagnosis has made 3D image reconstruction in radiological imaging a k

SaliencyDecor: Enhancing Neural Network Interpretability through Feature Decorrelation

arXiv:2604.25315v1 Announce Type: new Abstract: Gradient-based saliency methods are widely used to interpret deep neural networks, yet they often produce noisy and unstable explanations that poorly al

researcharxiv-cs-cv

SARU: A Shadow-Aware and Removal Unified Framework for Remote Sensing Images with New Benchmarks

arXiv:2604.25432v1 Announce Type: new Abstract: Shadows are a prevalent problem in remote sensing imagery (RSI), degrading visual quality and severely limiting the performance of downstream tasks like

Local Ai

SenseNova-U1: Unifying Multimodal Understanding and Generation with NEO-Unify Architecture

SenseNova U1 is a new series of native multimodal models that unifies multimodal understanding, reasoning, and generation within a monolithic architecture, marking a fundamental paradigm shift in mult

local-air-localllama

Shearlet Neural Operators for Anisotropic-Shock-Dominated and Multi-scale parametric partial differential equations

arXiv:2604.25181v1 Announce Type: new Abstract: Neural operators have emerged as powerful data-driven surrogates for learning solution operators of parametric partial differential equations (PDEs). Ho

SnapMLA: Efficient Long-Context MLA Decoding via Hardware-Aware FP8 Quantized Pipelining

arXiv:2602.10718v3 Announce Type: replace-cross Abstract: While FP8 attention has shown substantial promise in innovations like FlashAttention-3, its integration into the decoding phase of the DeepSee

So excited for this - in the past four months I've written all my code using Deep Agents with: - Opus 4.6 - GPT-5.4 - GLM-5 - Kimi 2.6 - Opu…

model-releasesharrison-chase--x

So excited for this - in the past four months I've written all my code using Deep Agents with: - Opus 4.6 - GPT-5.4 - GLM-5 - Kimi 2.6 - Opus 4.7 - GPT-5.5 Currently using 5.5 as my driver with Opus s

Soft-TransFormers for Continual Learning

arXiv:2411.16073v3 Announce Type: replace-cross Abstract: Inspired by the Well-initialized Lottery Ticket Hypothesis (WLTH), we introduce Soft-Transformer (Soft-TF), a parameter-efficient framework fo

Splatent: Splatting Diffusion Latents for Novel View Synthesis

arXiv:2512.09923v2 Announce Type: replace Abstract: Radiance field representations have recently been explored in the latent space of VAEs that are commonly used by diffusion models. This direction of

researcharxiv-cs-cv

Stay tuned for registration details https://openai.com/index/devday-2026/

OpenAI announced an upcoming DevDay 2026 event and indicated that registration details would be shared at a later time. The announcement was made via OpenAI's X (Twitter) account, directing interested

model-releasesopenai--x

Still wondering how you can use Codex for (almost) everything? Codex can help with more of the work that supports the work, from organizing …

Still wondering how you can use Codex for (almost) everything? Codex can help with more of the work that supports the work, from organizing research to making spreadsheets, decks, and summaries. Media

model-releasesopenai--x

StrADiff: A Structured Source-Wise Adaptive Diffusion Framework for Linear and Nonlinear Blind Source Separation

arXiv:2604.04973v2 Announce Type: replace-cross Abstract: This paper presents StrADiff, a Structured Source-Wise Adaptive Diffusion Framework for unsupervised blind source separation under linear and

researcharxiv-cs-lg

Subspace Optimization for Efficient Federated Learning under Heterogeneous Data

arXiv:2604.25467v1 Announce Type: new Abstract: Federated learning increasingly operates in a large-model regime where communication, memory, and computation are all scarce. Typically, non-IID client

researcharxiv-cs-lg

Task-Driven Prompt Learning: A Joint Framework for Multi-modal Cloud Removal and Segmentation

arXiv:2601.12052v2 Announce Type: replace Abstract: Optical remote sensing imagery is indispensable for Earth observation, yet persistent cloud occlusion limits its downstream utility. Most cloud remo

Teacher Forcing as Generalized Bayes: Optimization Geometry Mismatch in Switching Surrogates for Chaotic Dynamics

arXiv:2604.25904v1 Announce Type: new Abstract: Identity teacher forcing (ITF) enables stable training of deterministic recurrent surrogates for chaotic dynamical systems and has been highly effective

researcharxiv-cs-lg