Microsoft Research
EvoLib: Turning experience into evolving knowledge
EvoLib enables large language models to learn from their own experience during inference, without requiring…
ai · source
Microsoft Research on WeSearch
Recent ai headlines from Microsoft Research.
Lead story
Microsoft Research
Echoverse: Deep, evolving environments for computer-use agents
At a glance We built twelve training worlds for computer-use agents: ten deep domain worlds and two capability worlds, each drilling a single control rendered in many forms (date pickers and nested filters). Depth is…
The latest
Microsoft Research
Extending Human Intelligence Through AI
Modern AI systems enhance human intelligence by extending existing cognitive structures rather than replicating them.…
Microsoft Research
Microsoft Research: LLMs Corrupt your files during delegated work
A recent study reveals that large language models (LLMs) can corrupt documents during delegated tasks. The research,…
Microsoft Research
Multiple Byte Processing with Full-Word Instructions (1975)
The article discusses advancements in parallel computing techniques using multiple byte processing with full-word…
Microsoft Research
Webwright: A Terminal Is All You Need for Web Agents
Webwright introduces a new approach to web agents by utilizing a terminal-based setup instead of traditional browser…
Microsoft Research
MagenticLite, MagenticBrain, Fara1.5: An agentic experience optimized for small models
Microsoft Research has introduced MagenticLite, an agentic application optimized for small models. This new system…
Microsoft Research
Vega: Zero-knowledge proofs for digital identity in the age of AI
Vega introduces a method for users to prove aspects of their government-issued credentials without revealing the…
Microsoft Research
Guidelines for Human-AI Interaction
The Guidelines for Human-AI Interaction were established to improve the way AI systems interact with users. Developed…
How WeSearch handles this source
WeSearch's declared handling of Microsoft Research's content. Indexing, snippets, summaries, retrieval and training are separate questions — see the rights registry or read this source's machine-readable record.
Indexing: Allowed
Snippet: Allowed
AI summary: Limited
Retrieval / RAG: Not asserted
Model training: Not asserted
Commercial reuse: Not permitted