Research monitor · Page 3

Microsoft Research Focus – Edition #97

Agent systems · primary_technical_publication

Host Sowbhagya Suresh shares some of the latest news and cutting-edge research from across Microsoft—from understanding the challenges facing coding agents to uncovering how people

VeriCUA: Verification as Agentic Exploration for Computer-Using Agents

Agent systems · preprint

Building and scaling computer-using agents (CUAs) across desktop, mobile, and web environments requires reliable task-completion verification for data curation, evaluation, and reinforcement learning. Typical automated

Machine Learning-Based Detection of Mobile Money Scan Messages in Ghana: A Comparative Evaluation of Naive Bayes, Support Vector Machine, and Randon Forest Classifiers

Evaluation & safety · preprint

Mobile money (MoMo) is Ghana's preferred retail payment channel, recording GHC518.4 billion in transactions in December 2025 and over 26.7 million active accounts (Graphic Online,

From extreme weather to agricultural consequences with agentic AI

Agent systems · preprint

Extreme weather is forecast days ahead, yet its consequences for agriculture remain hard to predict: a forecast says where a storm will strike, not which

Ramanujan Challenge Results: AI-Assisted Evaluation Pipeline for Multi-Format Proofs

Mathematical reasoning · preprint

As AI's mathematical capabilities rapidly improve, meaningful benchmarks are shifting from university-level problems and math olympiads to research-level mathematics. The Ramanujan Challenge is a benchmark

Collaborative AI agents for autonomous experimental research with solid-state quantum spins

Agent systems · preprint

An autonomous research laboratory could conduct an entire study, from formulating questions to preparing manuscripts. Artificial intelligence (AI) offers one route towards this goal. Recent

BURGER: Benchmarking and Understanding Robotic Generalization via Embodied Drawing

Evaluation & safety · preprint

How effectively can vision-language-action (VLA) models learn diverse language-conditioned behaviors as training data scales? Investigating this question requires large and varied robot datasets, whose collection

TaskFrame: Connecting Task-Level Planning with Frame-Level Control for Game Agents

Agent systems · preprint

Building general game agents requires both high-level task reasoning and low-latency action control in rapidly changing visual environments. These capabilities favor models with substantially different

Learning Safe Flow Matching Robot Policies from Binary Feasibility Oracles

Robotics & embodiment · preprint

Robot foundation models increasingly rely on flow-based action generation for general-purpose manipulation, yet the generated actions are not guaranteed to satisfy physical constraints such as

Decision-Level Error Attribution and Depth-Aware Safety Evaluation of Monocular RGB Human Tracking for Autonomous Robot Navigation

Evaluation & safety · preprint

Monocular RGB tracking is useful for robots operating near people, but perception errors affect binary stop/proceed decisions in ways that geometric accuracy metrics do not

Beyond Selected Breakthroughs: Evaluating Language Models on Randomly Sampled Mathematical Conjectures

Mathematical reasoning · preprint

The last year has seen a shift in AI-assisted mathematics, with frontier language models producing proofs of and counterexamples to long-standing open conjectures. However, studying

RoboSeed: Multidimensional Evaluation of Vision-Language-Action and World-Action Models: Capability, Resource Affordability, and Research Evidence

Evaluation & safety · preprint

Choosing a robot action model requires evidence about task capability, adaptation cost, deployment requirements, and the reliability of reported conclusions. Success rates alone offer limited

SwapBot: Tool-Centric Robot Manipulation via Interchangeable End-Effectors and Tactile Sensing

Robotics & embodiment · preprint

Robot manipulation in everyday environments inherently involves substantial tool usage. Seemingly simple household tasks, such as cooking pan-fried tofu, require forceful cutting, delicate brushing, and

H$^2$R: Hierarchical Hint-Based Reward for Mathematical Reasoning via Reinforcement Learning

Mathematical reasoning · preprint

Reinforcement Learning with Verifiable Rewards has shown great promise in enhancing mathematical reasoning in large language models. However, existing approaches provide limited intermediate supervision for

How we detect frontier model misalignment

Agent systems · provider_announcement

OpenAI uses layered monitoring to detect frontier model misalignment and unauthorized agent behavior so responders can intervene before incidents escalate.

EpiAgent: Epistemic Self-Play for LLM Agents

Agent systems · preprint

Epistemic game theory explains an agent's strategic behavior via the beliefs it holds about others and how these beliefs are updated as the interactions unfold.

MUSBRIC: LEGO-Compatible Pneumatic Muscle Bricks for Reconfigurable Musculoskeletal Robots

Robotics & embodiment · preprint

Musculoskeletal robots combine compliant actuators with articulated skeletons, but exploring alternative configurations often requires repeated fabrication. We present MUSBRIC, a modular hardware platform and accompanying

AI for Science expands to foundations and institutions

AI for science · provider_announcement

We’re highlighting 20 scientific foundations, institutions, and research groups from around the world that we’ve partnered with in an expansion of our AI for Science

The future of the firm

AI models · provider_announcement

How AI might enter the org chart—beneath, beside, and between us.

OpenAI and ServiceNow expand partnership across enterprise workflows

Agent systems · provider_announcement

OpenAI and ServiceNow expand their partnership to bring OpenAI models and agents into ServiceNow and ServiceNow workflows into ChatGPT and Codex.

Legora is building the agentic operating system for legal work

Agent systems · provider_announcement

See how Legora uses OpenAI models, ChatGPT, and Codex to accelerate legal workflows, expand contract analysis, and free lawyers to focus on higher-value work.

Gemini 4 Carbon: Our next frontier model

AI models · provider_announcement

Genie 4: A new generation of world simulation

Robotics & embodiment · provider_announcement

Genie 4: our new world model unlocking unprecedented realism in real-time simulation.

OpenAI expands partnership with Oracle to bring its apps to ChatGPT

Agent systems · provider_announcement

OpenAI and Oracle are expanding their partnership to bring business applications and AI agents into ChatGPT and power new Oracle offerings.