model releases

Today we're announcing ESMFold2, an open scientific engine to power prediction, design, and discovery across protein biology. The new model …

model-releasesyann-lecun--x

Today we're announcing ESMFold2, an open scientific engine to power prediction, design, and discovery across protein biology. The new model delivers state of the art performance on protein interaction

Model Releases

Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets

DGX agent

arXiv:2605.26165v1 Announce Type: cross Abstract: Agentic RAG systems that equip language models with dozens to hundreds of tool definitions face a critical resource conflict: tool schemas consume the

model-releasesarxiv-cs-ai

27 May 2026

Towards Error-Free EHRs: Reasoning-Intensive Consistency Verification Between Clinical Notes and Structured Tables in Electronic Health Records

arXiv:2605.26463v1 Announce Type: cross Abstract: Data consistency between unstructured clinical notes and structured tables in Electronic Health Records (EHRs) is essential for patient safety and cli

Towards Generalization-Oriented Models for Vehicle Routing Problems with Mixture-of-Experts

arXiv:2605.26776v1 Announce Type: cross Abstract: In recent years, Deep Reinforcement Learning (DRL) has achieved substantial progress on Vehicle Routing Problems (VRPs). However, existing DRL-based m

TowerMind: A Tower Defence Game Learning Environment and Benchmark for LLM as Agents

arXiv:2601.05899v2 Announce Type: replace Abstract: Recent breakthroughs in Large Language Models (LLMs) have positioned them as a promising paradigm for agents, with long-term planning and decision-m

Traceable Knowledge Graph Reasoning Enables LLM-Assisted Decision Support for Industrial VOCs in the Steel Industry

arXiv:2605.27071v1 Announce Type: new Abstract: Key knowledge for steel-industry volatile organic compounds (VOCs) governance is scattered across unstructured scientific literature, making it difficul

Trust Region Q Adjoint Matching

arXiv:2605.27079v1 Announce Type: cross Abstract: Off-policy reinforcement learning of pretrained flow policies remains challenging due to the instability of optimization arising from the multi-step s

Two-Parameter Flows for Learning Population Dynamics of Physical Systems

arXiv:2605.26285v1 Announce Type: new Abstract: This work addresses the problem of learning the dynamics of high-dimensional probability densities over time using unlabeled samples, without assuming a

Underwater360: Reconstructing Underwater Scenes from Panoramic Images with Omnidirectional Gaussian Splatting

model-releasesarxiv-cs-cv

arXiv:2605.26447v1 Announce Type: new Abstract: Underwater scene reconstruction is essential for immersive exploration of aquatic environments, yet remains challenging due to complex participating-med

UnityMAS-O: A General RL Optimization Framework for LLM-Based Multi-Agent Systems

arXiv:2605.26646v1 Announce Type: new Abstract: LLM-based multi-agent systems decompose complex tasks into interacting roles, but most remain manually orchestrated by prompts, tools, and control rules

Variational Inference for Evidential Deep Learning

arXiv:2605.26477v1 Announce Type: new Abstract: While Deep Neural Networks (DNNs) achieve remarkable performance, their tendency to produce overconfident predictions. Evidential Deep Learning (EDL) mi

Vectors Are Not Neutral: Sensitive-Information Inference from Exported LLM Representations in Summarization

arXiv:2605.26433v1 Announce Type: new Abstract: Large language model (LLM) summarization systems may pass compact vector representations of private inputs to downstream retrieval, monitoring, audit, o

Verus-SpecGym: An Agentic Environment for Evaluating Specification Autoformalization

arXiv:2605.26457v1 Announce Type: cross Abstract: AI coding agents are increasingly used to write real-world software, but ensuring that their outputs are correct remains a fundamental challenge. Form

VISTA: An End-to-End Benchmark for Visual Spec-to-Web-App Coding Agents

arXiv:2605.26144v1 Announce Type: cross Abstract: We present VISTA (VIsual Spec-To-App Benchmark), a benchmark for evaluating the end-to-end web-app generation capabilities of LLM-based agents. Unlike

VisualNeedle: Benchmarking Active Visual Search in Information-Dense Scenes

arXiv:2605.26380v1 Announce Type: cross Abstract: Frontier multimodal large language models (MLLMs) have been reported to achieve over 90% accuracy on fine-grained perception benchmarks. However, such

VitaBench 2.0: Evaluating Personalized and Proactive Agents in Long-Term User Interactions

arXiv:2605.27141v1 Announce Type: new Abstract: Large language models (LLMs) have evolved into interactive agents that collaborate with users in real-world tasks. Effective collaboration in such setti

Warp’s big bet on building open source with GPT-5.5

Warp is making a significant investment in developing open source tools and integrations built on GPT-5.5, OpenAI's advanced language model. The initiative aims to leverage GPT-5.5's capabilities to c

model-releasesopenai

We’ve been putting a lot of effort into making Claude Code more responsive & reliable. Here’s an update on everything we’ve done:

Anthropic has made improvements to Claude Code's responsiveness and reliability, with Thariq providing details on the enhancements implemented. The update likely covers performance optimizations, bug

model-releasesthariq--x

We've created the world's fastest PDF parser ⚡️ And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf,…

model-releasesjerry-liu--x

We've created the world's fastest PDF parser ⚡️ And it's more accurate than any other open-source, model-free PDF parser out there (pymupdf, pypdf, markitdown, pdftotext, opendataloader, pymupdf4llm)

What Demands Attention in Urban Street Scenes? From Scene Understanding towards Road Safety: A Survey of Vision-driven Datasets and Studies

model-releasesarxiv-cs-cv

arXiv:2507.06513v3 Announce Type: replace Abstract: Advances in vision-based sensors and computer vision algorithms have significantly improved the analysis and understanding of traffic scenarios. To

What Makes Chain-of-Thought Work at Probe Time? Local Co-occurrence Rather Than Global Derivation

arXiv:2605.26795v1 Announce Type: new Abstract: Chain-of-thought (CoT) prompting reliably improves language-model accuracy, but which properties of a rationale text drive the improvement is poorly und

What Molecular Structure Cannot Tell Us: A Taxonomy of Explainability Gaps in GNN-Based Drug Toxicity Prediction

arXiv:2605.26183v1 Announce Type: cross Abstract: Graph Neural Networks (GNNs) have emerged as a structurally natural approach for molecular toxicity prediction, operating directly on atomic connectiv

When Does Deep RL Beat Calibrated Baselines? A Benchmark Study on Adaptive Resource Control

arXiv:2605.26418v1 Announce Type: cross Abstract: A properly calibrated rule-based autoscaler can beat every one of six mainstream deep reinforcement learning (DRL) algorithms on cost across every wor

When LLMs Benchmark Themselves: Deconstructing Self-Bias in Automated Evaluation

arXiv:2509.26600v2 Announce Type: replace-cross Abstract: As LLMs rapidly saturate existing benchmarks, automated benchmark creation using LLMs (LLM-as-a-benchmark) -- where a model generates test inp

Why Prompt Optimization Works, and Why It Sometimes Doesn't: A Causal-Inspired Edit-Level Analysis

arXiv:2605.26655v1 Announce Type: new Abstract: Automated prompt optimization methods (e.g., DSpy, TextGrad) can substantially improve the performance of large language model (LLM), however, their gen

WINDQuant: Weight-Informed Neural Decision-Making for Global Mixed-Precision LLM Quantization

arXiv:2605.26660v1 Announce Type: new Abstract: Quantization is an effective approach to reduce the memory footprint and inference cost of large language models (LLMs), yet maintaining performance in

Workflow Closure Is Not Scientific Closure in Auto-Research Systems

arXiv:2605.26200v1 Announce Type: cross Abstract: This paper argues that workflow closure is not scientific closure in auto-research systems. Current systems can increasingly complete research-like lo

Xreal launches a new sub-brand, X by Xreal, and the $299 a01 display glasses with micro OLED displays, 50° FOV, and 62g weight, set for July release in the US (Scott Stein/CNET)

Scott Stein / CNET: Xreal launches a new sub-brand, X by Xreal, and the 299 a01 display glasses with micro OLED displays, 50° FOV, and 62g weight, set for July release in the US — X by Xreal is arrivi

model-releasestechmeme

Your Agents Are Aging Too: Agent Lifespan Engineering for Deployed Systems

arXiv:2605.26302v1 Announce Type: new Abstract: Long-lived AI agents are increasingly deployed as persistent operational systems, yet they are still evaluated like freshly initialized models. Day-one

// Your Agents are Aging Too // Huh!? They need 'sleep,' and now they are aging? Joke aside, great write-up on reliable agentic engineering.…

// Your Agents are Aging Too // Huh!? They need 'sleep,' and now they are aging? Joke aside, great write-up on reliable agentic engineering. This new research introduces AgingBench, a longitudinal rel

model-releasesdair-ai--x

Zero-Shot MARL Benchmark in the Cyber-Physical Mobility Lab

model-releasesarxiv-cs-ro

arXiv:2601.16578v2 Announce Type: replace Abstract: We present a reproducible benchmark for evaluating sim-to-real transfer of Multi-Agent Reinforcement Learning (MARL) policies for Connected and Auto

Zero-Shot Object Re-Identification in Egocentric Kitchen Videos via Multi-Stage SAM3 Feature Fusion

model-releasesarxiv-cs-cv

arXiv:2605.26383v1 Announce Type: new Abstract: Object re-identification (ReID) in egocentric kitchen videos is challenging due to rapid viewpoint changes, frequent occlusions, cluttered scenes, and l

7AI launches PLAID ELITE fully managed agentic security operations service

model-releasessiliconangle

Agentic artificial intelligence security startup 7AI Inc. today announced the launch of PLAID ELITE, a fully managed AI-native security operations service. The new service combines autonomous investig

A Blended Likelihood Approach for Achieving Fairness Using Naive Bayes

arXiv:2605.25228v1 Announce Type: new Abstract: Concerns about algorithmic bias and fairness have increased as artificial intelligence has been incorporated into high-stakes decision-making. Tradition

A Comprehensive Dataset for Human vs. AI Generated Text Detection

arXiv:2510.22874v3 Announce Type: replace Abstract: The rapid advancement of large language models (LLMs) has led to increasingly human-like AI-generated text, raising concerns about content authentic

A computational phase transition for learning-to-sample from Ising models

arXiv:2605.24752v1 Announce Type: new Abstract: We study learning-to-sample -- a basic algorithmic task underlying generative modeling -- for Ising models, a standard testbed for algorithmic ideas in

A Controlled Synthetic Benchmark for Educational Aspect-Based Sentiment Analysis

arXiv:2605.25502v1 Announce Type: cross Abstract: Educational aspect-based sentiment analysis (ABSA) can support course improvement, but public aspect-labeled student feedback remains scarce because e

A Large-Scale Dataset and Benchmark: Do Protein-Ligand Models Learn Binding Sites or Just Binding Likelihood?

arXiv:2605.24045v1 Announce Type: cross Abstract: Protein-ligand modeling underpins computational drug discovery and molecular design. Existing protein-ligand benchmarks typically evaluate whether a p

A Learning Stability Profile for Finite-Dimensional Learning Dynamics

arXiv:2512.21208v3 Announce Type: replace Abstract: We develop a finite-dimensional sensitivity framework for studying stability in learning systems whose states include representations, parameters, a

A lift for input-convex neural network training

arXiv:2605.24274v1 Announce Type: new Abstract: Input-convex neural networks (ICNNs) are widely used for log-concave density estimation, convex-potential normalizing flows, optimal transport, and tran

A Matched Spectral Benchmark of Quantum Inspired Feature Maps

arXiv:2605.24324v1 Announce Type: cross Abstract: Quantum machine learning is often motivated by the idea that quantum systems can expose useful high-dimensional structure that is difficult to access

A Multi-Probe Audit of Clinical-Interview Depression Detection Benchmarks

arXiv:2605.23977v1 Announce Type: new Abstract: This paper audits benchmark evaluation in clinical-interview depression detection through four complementary probes across DAIC/E-DAIC, CMDC, ANDROIDS,

A Tabular Schedule Abstraction for Communication-Aware Evaluation of Pipeline-Parallel LLM Training

arXiv:2605.24006v1 Announce Type: cross Abstract: Pipeline parallelism is a key technique for distributed training of large language models because it reduces per-device parameter and activation memor

A tool to get Claude Code-style reliability from fully local models

Ollama exposes an Anthropic-compatible Messages endpoint , allowing developers to run powerful open-source AI models locally with no API costs and pair them with Claude Code for a capable local AI cod

model-releasesr-ollama

A Two-Phase Stability Study of LLM Judges and Bar Council Examiners on Thai Bar-Exam Free-Form Essays

arXiv:2605.25652v1 Announce Type: new Abstract: Free-form legal essay evaluation in NLP treats expert inter-rater stability as a single ceiling number, and treats LLM-judge agreement with that ceiling

A World Model of Radiologist Reading for Medical Image Representation Learning

arXiv:2605.23992v1 Announce Type: cross Abstract: Radiologist eye-tracking data provide a rich record of how experts search, compare, and accumulate evidence during image reading; yet, existing method

Acting on the Unseen: Communication-Free Collaborative Filtering for Decentralized Multi-Robot Task Allocation

arXiv:2605.25584v1 Announce Type: cross Abstract: Multi-robot task allocation usually assumes some combination of communication, known task models, or a coordinator. We study the opposite extreme, a r

Action-Prior Denoising for Smooth Real-Time Chunking

model-releasesarxiv-cs-ro

arXiv:2605.25537v1 Announce Type: new Abstract: Real-time chunking (RTC) lets chunked action policies operate under inference delay by conditioning a newly generated action chunk on actions already co