Differential Transformer V2
• 52
AI & ML interests
None defined yet.
Recent Activity
View all activity
Papers
FastContext: Training Efficient Repository Explorer for Coding Agents
A Benchmark and Framework for Evaluating Next Action Predictions in Spreadsheets
Articles
SWE-FastContext
A family of code-search models powering the Explore subagent for coding agents.
GridSFM
Collection of datasets and models developed to support research in power grid modeling
ChatBench
ChatBench Datasets and Simulators (same prompt + fine-tuning set-up) from the ChatBench paper.
MediPhi
A collection of SLMs based on Phi3.5-mini-instruct adapted to clinical natural language processing tasks: https://arxiv.org/abs/2505.10717
NatureLM
NextCoder
NextCoder family of code-editing LMs developed with Selective Knowledge Transfer and its training data.
Phi-3
Phi-3 family of small language and multi-modal models. Language models are available in short- and long-context lengths.
Controllable Safety Alignment
Artifacts for the paper "Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements" (https://arxiv.org/abs/2410.08968)
MAI-DS-R1
MAI-DS-R1 is a DeepSeek-R1 reasoning model that has been post-trained by the Microsoft AI team.
SpeechT5
The SpeechT5 framework consists of a shared seq2seq and six modal-specific (speech/text) pre/post-nets that can address a few audio-related tasks.
Table Transformer
The Table Transformer (TATR) is a series of object detection models useful for table extraction from PDF images.
Biomedical
Models for biomedical research applications, such as radiology report generation and biomedical language understanding.
UDOP
UDOP is a general multimodal model for document AI
Florence
MoCapAct
Locomotion policies for hundreds of simulated humanoid locomotion clips and demonstration data for training them.
Froggy-Models
Skala
Accurate and scalable exchange-correlation with deep learning
VibeVoice
Frontier Text-to-Speech Models https://microsoft.github.io/VibeVoice/
Dayhoff Atlas
The models and datasets that comprise the Dayhoff Atlas
Paza
Paza is a collection of speech models & benchmarks for low resource languages by the Microsoft Research Africa - Nairobi Lab
Phi-4
Phi-4 family of small language, multi-modal and reasoning models.
Phi-1
Phi-1 family of small language models.
BitNet
🔥BitNet family of large language models (1-bit LLMs).
LLM2CLIP
LLM2CLIP makes SOTA pretrained CLIP modal more SOTA ever.
TAPEX
TAPEX is the state-of-the-art table pre-training models which can be used for table-based question answering and table-based fact verification.
LayoutLM
The LayoutLM series are Transformer encoders useful for document AI tasks such as invoice parsing, document image classification and DocVQA.
Orca
The Orca family of LMs developed by Microsoft.
GIT
GIT (Generative Image-to-text Transformer) is a model useful for vision-language tasks such as image/video captioning and question answering.
IFMs
Industrial Foundation Models
SWE-FastContext
A family of code-search models powering the Explore subagent for coding agents.
Froggy-Models
GridSFM
Collection of datasets and models developed to support research in power grid modeling
Skala
Accurate and scalable exchange-correlation with deep learning
ChatBench
ChatBench Datasets and Simulators (same prompt + fine-tuning set-up) from the ChatBench paper.
VibeVoice
Frontier Text-to-Speech Models https://microsoft.github.io/VibeVoice/
MediPhi
A collection of SLMs based on Phi3.5-mini-instruct adapted to clinical natural language processing tasks: https://arxiv.org/abs/2505.10717
Dayhoff Atlas
The models and datasets that comprise the Dayhoff Atlas
NatureLM
Paza
Paza is a collection of speech models & benchmarks for low resource languages by the Microsoft Research Africa - Nairobi Lab
NextCoder
NextCoder family of code-editing LMs developed with Selective Knowledge Transfer and its training data.
Phi-4
Phi-4 family of small language, multi-modal and reasoning models.
Phi-3
Phi-3 family of small language and multi-modal models. Language models are available in short- and long-context lengths.
Phi-1
Phi-1 family of small language models.
Controllable Safety Alignment
Artifacts for the paper "Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements" (https://arxiv.org/abs/2410.08968)
BitNet
🔥BitNet family of large language models (1-bit LLMs).
MAI-DS-R1
MAI-DS-R1 is a DeepSeek-R1 reasoning model that has been post-trained by the Microsoft AI team.
LLM2CLIP
LLM2CLIP makes SOTA pretrained CLIP modal more SOTA ever.
SpeechT5
The SpeechT5 framework consists of a shared seq2seq and six modal-specific (speech/text) pre/post-nets that can address a few audio-related tasks.
TAPEX
TAPEX is the state-of-the-art table pre-training models which can be used for table-based question answering and table-based fact verification.
Table Transformer
The Table Transformer (TATR) is a series of object detection models useful for table extraction from PDF images.
LayoutLM
The LayoutLM series are Transformer encoders useful for document AI tasks such as invoice parsing, document image classification and DocVQA.
Biomedical
Models for biomedical research applications, such as radiology report generation and biomedical language understanding.
Orca
The Orca family of LMs developed by Microsoft.
UDOP
UDOP is a general multimodal model for document AI
GIT
GIT (Generative Image-to-text Transformer) is a model useful for vision-language tasks such as image/video captioning and question answering.
Florence
IFMs
Industrial Foundation Models
MoCapAct
Locomotion policies for hundreds of simulated humanoid locomotion clips and demonstration data for training them.