Recherche
Discover an index of datasets, SDKs, APIs and open-source tools developed by Microsoft researchers and shared with the global academic community below. These experimental technologies—available through Azure AI Foundry Labs (opens in new tab)—offer a glimpse into the future of AI innovation.
Structure-aware Multi-Objective Metaprompt Optimization Library for Python. A flexible, easy-to-use library for running and optimizing prompts for Large Language Models (LLMs).
CodePlan is a research project that formalizes repository-level coding tasks as planning problems and uses static analysis and large language models (LLMs) to solve them. This replication package is for the paper titled “CodePlan: Repository-level…
Algorithms to Handle Signal Delay in Deep Reinforcement Learning aims to address the problem of signal delay in continuous robotic control. Signal delay occurs when there is a lag between an agent’s perception of the…
A large-scale egocentric human interaction dataset, where two people collaboratively complete physical manipulation tasks.
Orca 2 is a finetuned version of LLAMA-2. It is built for research purposes only and provides a single turn response in tasks such as reasoning over user given data, reading comprehension, math problem solving…
Orca 2 is a finetuned version of LLAMA-2. It is built for research purposes only and provides a single turn response in tasks such as reasoning over user given data, reading comprehension, math problem solving…
LLF Bench is a benchmark for evaluating learning agents that provides a diverse collection of interactive learning problems where the agent gets language feedback instead of rewards (as in RL) or action feedback (as in…
This repo contains data and code for EMNLP’23 paper “Fifty Shades of Bias”: Normative Ratings of Gender Bias in GPT Generated English Text
Tyger is a framework for remote signal processing. It enables reliable transmission of data to remote computational resources, where the data can be processed and transformed as it streams in. It was designed for streaming…
This work aims at decreasing the end-to-end generation latency of large language models (LLMs). One of the major causes of the high generation latency is the sequential decoding approach adopted by almost all state-of-the-art LLMs.…