Microsoft

company

Verified

https://www.microsoft.com/en-us/research/

microsoft

AI & ML interests

None defined yet.

Recent Activity

frontierai updated a model 36 minutes ago

microsoft/VibeVoice-AcousticTokenizer

frontierai published a model 37 minutes ago

microsoft/VibeVoice-AcousticTokenizer

Baolin submitted a paper about 1 hour ago

Reinforcement World Model Learning for LLM-based Agents

View all activity

Papers

RE-TRAC: REcursive TRAjectory Compression for Deep Search Agents

PISCES: Annotation-free Text-to-Video Post-Training via Optimal Transport-Aligned Rewards

View all Papers

Articles

Differential Transformer V2

Introducing OptiMind, a research model designed for optimization

microsoft 's Papers 28

Submitted by

Jialiang Zhu

RE-TRAC: REcursive TRAjectory Compression for Deep Search Agents

microsoft

Submitted by

Minh-Quan Le

PISCES: Annotation-free Text-to-Video Post-Training via Optimal Transport-Aligned Rewards

microsoft

Submitted by

Mingqian Feng

Statistical Estimation of Adversarial Risk in Large Language Models under Best-of-N Sampling

microsoft

Submitted by

Tianyi Chen

CUA-Skill: Develop Skills for Computer Using Agent

microsoft

Submitted by

Furu Wei

VIBEVOICE-ASR Technical Report

microsoft

Submitted by

taesiri

WebGym: Scaling Training Environments for Visual Web Agents with Realistic Tasks

microsoft

Submitted by

Jue Zhang

DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems

microsoft

Submitted by

Xiao Liang

Gold-Medal-Level Olympiad Geometry Solving with Efficient Heuristic Auxiliary Constructions

microsoft

Submitted by

taesiri

Fara-7B: An Efficient Agentic Model for Computer Use

microsoft

Submitted by

Chaoyun Zhang

UFO^3: Weaving the Digital Agent Galaxy

microsoft

Submitted by

Chaoyun Zhang

GUI-360: A Comprehensive Dataset and Benchmark for Computer-Using Agents

microsoft

Submitted by

Huanyu_Zhang

Latent Sketchpad: Sketching Visual Thoughts to Elicit Multimodal Reasoning in MLLMs

microsoft

Submitted by

Jannis Vamvas

QueST: Incentivizing LLMs to Generate Difficult Problems

microsoft

3

Submitted by

Shun Zheng

Deep Self-Evolving Reasoning

microsoft

2

Submitted by

Jiayu Ding

Information-Preserving Reformulation of Reasoning Traces for Antidistillation

microsoft

Submitted by

taesiri

SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs

microsoft

Submitted by

taesiri

Just Do It!? Computer-Use Agents Exhibit Blind Goal-Directedness

microsoft

3

Submitted by

Minki Kang

ACON: Optimizing Context Compression for Long-horizon LLM Agents

microsoft

Submitted by

Miaosen Zhang

InfoAgent: Advancing Autonomous Information-Seeking Agents

microsoft

2

Submitted by

Pranjal A. Chitale

The role of synthetic data in Multilingual, Multi-cultural AI systems: Lessons from Indic Languages

microsoft

2

Submitted by

Ruiyu Wang

CAD-Tokenizer: Towards Text-based CAD Prototyping via Modality-Specific Tokenization

microsoft

2

Submitted by

Xiao Liu

Behind RoPE: How Does Causal Mask Encode Positional Information?

microsoft

Submitted by

Eric Lan

Contextual Integrity in LLMs via Reasoning and Reinforcement Learning

microsoft

TwT: Thinking without Tokens by Habitual Reasoning Distillation with Multi-Teachers' Guidance

microsoft

PromptIntern: Saving Inference Costs by Internalizing Recurrent Prompt during Large Language Model Fine-tuning

microsoft

Submitted by

AK

Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

microsoft

Submitted by

AK

WizardCoder: Empowering Code Large Language Models with Evol-Instruct

microsoft

Table Meets LLM: Can Large Language Models Understand Structured Table Data? A Benchmark and Empirical Study

microsoft

microsoft (Microsoft)

Microsoft

company

Verified

https://www.microsoft.com/en-us/research/

microsoft

AI & ML interests

None defined yet.

Recent Activity

frontierai updated a model 36 minutes ago

microsoft/VibeVoice-AcousticTokenizer

frontierai published a model 37 minutes ago

microsoft/VibeVoice-AcousticTokenizer

Baolin submitted a paper about 1 hour ago

Reinforcement World Model Learning for LLM-based Agents

View all activity

Papers

RE-TRAC: REcursive TRAjectory Compression for Deep Search Agents

PISCES: Annotation-free Text-to-Video Post-Training via Optimal Transport-Aligned Rewards

View all Papers

Articles

Differential Transformer V2

Introducing OptiMind, a research model designed for optimization

microsoft 's Papers 28

Submitted by

Jialiang Zhu

RE-TRAC: REcursive TRAjectory Compression for Deep Search Agents

microsoft

Submitted by

Minh-Quan Le

PISCES: Annotation-free Text-to-Video Post-Training via Optimal Transport-Aligned Rewards

microsoft

Submitted by

Mingqian Feng

Statistical Estimation of Adversarial Risk in Large Language Models under Best-of-N Sampling

microsoft

Submitted by

Tianyi Chen

CUA-Skill: Develop Skills for Computer Using Agent

microsoft

Submitted by

Furu Wei

VIBEVOICE-ASR Technical Report

microsoft

Submitted by

taesiri

WebGym: Scaling Training Environments for Visual Web Agents with Realistic Tasks

microsoft

Submitted by

Jue Zhang

DoVer: Intervention-Driven Auto Debugging for LLM Multi-Agent Systems

microsoft

Submitted by

Xiao Liang

Gold-Medal-Level Olympiad Geometry Solving with Efficient Heuristic Auxiliary Constructions

microsoft

Submitted by

taesiri

Fara-7B: An Efficient Agentic Model for Computer Use

microsoft

Submitted by

Chaoyun Zhang

UFO^3: Weaving the Digital Agent Galaxy

microsoft

Submitted by

Chaoyun Zhang

GUI-360: A Comprehensive Dataset and Benchmark for Computer-Using Agents

microsoft

Submitted by

Huanyu_Zhang

Latent Sketchpad: Sketching Visual Thoughts to Elicit Multimodal Reasoning in MLLMs

microsoft

Submitted by

Jannis Vamvas

QueST: Incentivizing LLMs to Generate Difficult Problems

microsoft

3

Submitted by

Shun Zheng

Deep Self-Evolving Reasoning

microsoft

2

Submitted by

Jiayu Ding

Information-Preserving Reformulation of Reasoning Traces for Antidistillation

microsoft

Submitted by

taesiri

SwiReasoning: Switch-Thinking in Latent and Explicit for Pareto-Superior Reasoning LLMs

microsoft

Submitted by

taesiri

Just Do It!? Computer-Use Agents Exhibit Blind Goal-Directedness

microsoft

3

Submitted by

Minki Kang

ACON: Optimizing Context Compression for Long-horizon LLM Agents

microsoft

Submitted by

Miaosen Zhang

InfoAgent: Advancing Autonomous Information-Seeking Agents

microsoft

2

Submitted by

Pranjal A. Chitale

The role of synthetic data in Multilingual, Multi-cultural AI systems: Lessons from Indic Languages

microsoft

2

Submitted by

Ruiyu Wang

CAD-Tokenizer: Towards Text-based CAD Prototyping via Modality-Specific Tokenization

microsoft

2

Submitted by

Xiao Liu

Behind RoPE: How Does Causal Mask Encode Positional Information?

microsoft

Submitted by

Eric Lan

Contextual Integrity in LLMs via Reasoning and Reinforcement Learning

microsoft

TwT: Thinking without Tokens by Habitual Reasoning Distillation with Multi-Teachers' Guidance

microsoft

PromptIntern: Saving Inference Costs by Internalizing Recurrent Prompt during Large Language Model Fine-tuning

microsoft

Submitted by

AK

Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone

microsoft

Submitted by

AK

WizardCoder: Empowering Code Large Language Models with Evol-Instruct

microsoft

Table Meets LLM: Can Large Language Models Understand Structured Table Data? A Benchmark and Empirical Study

microsoft