Posts in category: Microsoft AI

This research paper was presented at the 64^th IEEE Symposium on Foundations of Computer Science (FOCS) 2023 (opens in new tab), a premier forum for the latest research in theoretical computer science.

Submodular functions are versatile mathematical tools, finding diverse applications in real-world scenarios and guiding solutions across complex domains. From dissecting the intricate networks of graphs to deciphering the complexities of economic landscapes through utility functions, and even navigating the enigmatic world of random variables via entropy functions, they offer valuable insights into challenging problems. Their wide-ranging applicability has made them pivotal tools for modeling and optimization in various theoretical computer science domains, including operations research and game theory. In recent years, submodular functions have gained prominence in solving optimization problems within machine learning (ML) applications. These tasks encompass vital areas such as feature selection and clustering, as illustrated in Figure 1. Additionally, submodular functions are instrumental in applications like sensor placement and graphical models. For further exploration, comprehensive resources are available in Bilmes’ insightful survey (opens in new tab) and Bach’s standard textbook (opens in new tab) on this subject.

Algorithm design for submodular function minimization

In a joint paper with researchers from Stanford University, “Sparse Submodular Function Minimization(opens in new tab) (opens in new tab),” presented at FOCS 2023(opens in new tab) (opens in new tab), we investigate the problem of minimizing a submodular function in the standard model.   Here, we assume that the submodular function can be accessed through an evaluation oracle that returns the value ( f(S) ) in response to a query with a set ( S ). This is the most classical and well-studied model for studying algorithm design for minimizing submodular functions.

Before we discuss our study, it’s important to bear in mind that a submodular function ( f ) is defined on subsets of a finite set of elements ( V ) that satisfy a diminishing marginal difference property. That is, for any two subsets ( S subseteq T ) and any element ( e in V setminus T ), the marginal value of ( e ) when added to the smaller set ( f(S cup {e}) – f(S) ) is at least the marginal value of ( e ) when added to the bigger set ( f(T cup {e}) – f(T) ).

In the 1980s, foundational work (opens in new tab) revealed that submodular functions could be minimized in polynomial time, marking a significant breakthrough. Since then, researchers have made substantial progress in the quest for faster algorithms for submodular function minimization (SFM). Despite these efforts, fundamental questions persist, such as determining the minimum number of queries required to minimize any given submodular function—a concept referred to as the problem’s query complexity.

Currently, the most advanced algorithm needs to make ( widetilde{O}(n^2) ) queries for any given submodular function, while the best lower bound is only ( widetilde{Omega}(n) ), where (n) is the size of the ground set on which the submodular function is defined. This disparity results in a substantial gap, leaving an (n)-fold difference between the existing upper and lower bounds.

Given this considerable difference, a natural question arises: What additional structural assumptions could potentially pave the way for faster algorithms in submodular function minimization (SFM)? One prevalent assumption is sparsity, which posits that the size of the set minimizing the submodular function is small. This holds particular relevance in diverse applications, including signal processing, feature selection, and compressed sensing. In these scenarios, solutions are expected to exhibit sparse non-zero entries, making it important to understand how algorithmic complexity depends on sparsity, as it provides insights into the intricate combinatorial and geometric structures of the problems.

Interestingly, existing algorithmic techniques developed over the past four decades for SFM do not yield improved runtimes even when the solution is sparse. Therefore, it is imperative to develop innovative techniques that can drive advancements in sparse SFM and bridge the existing gap between upper and lower bounds.

Microsoft Research Podcast

AI Frontiers: The future of causal reasoning with Emre Kiciman and Amit Sharma

Emre Kiciman and Amit Sharma discuss their paper “Causal Reasoning and Large Language Models: Opening a New Frontier for Causality” and how it examines the causal capabilities of large language models (LLMs) and their implications.

Listen now

Opens in a new tab

Parallel algorithms for submodular function minimization

Exploring beyond SFM’s query complexity, recent research has shed light on the importance of sparse SFM, particularly in understanding the inherent adaptivity of parallel algorithms (known as parallel complexity) designed to solve the problem. Research has shown that any parallel algorithm for SFM requires a minimum adaptivity that is a polynomial in the size of the ground set.

Our results improve both parallel and sequential algorithms for SFM. For example, consider a scenario where the minimizer of the given submodular function is (widetilde{O}(1))-sparse. In this context, our parallel algorithm runs in a nearly constant number of rounds, while our sequential algorithm makes a nearly linear number of queries. This achievement stands in stark contrast with the previous best parallel upper bound of (widetilde{O}(n)) and the best query complexity upper bound of (widetilde{O}(n^2)).

Fast first-order methods for exact submodular function minimization

Current fast algorithms for SFM rely on cutting-plane methods, a standard class of convex optimization techniques applied to the Lovász extension—a natural continuous extension of the given submodular function. However, restricting the optimization domain to sparse solutions doesn’t significantly expedite cutting-plane methods beyond a logarithmic factor. To address this, we shifted our approach and employed first-order methods, including stochastic mirror descent, to minimize the Lovász extension. These methods, non-Euclidean generalizations of stochastic gradient descent, are more attuned to problem geometry. Unlike cutting-plane methods, first-order methods exhibit a polynomial convergence rate, rather than a polylogarithmic dependency on the additive error concerning the optimal solution. 

This rate of convergence indicates that first-order methods are better suited for approximate submodular function minimization, while our goal is to solve it exactly. Using the sparsity assumption, we developed a new algorithmic framework for SFM based on a new concept of duality. We used this framework to demonstrate how first-order methods, with substantially reduced accuracy requirements, can be applied to solve SFM exactly.

Toward faster algorithms for SFM and its applications

These techniques not only promise advancements for sparse SFM but also provide a foundation for tackling other fundamental problems in SFM theory. Our algorithms for sparse SFM serve as valuable starting points for designing improved algorithms for related problems. They offer potential insights into developing polynomial-time algorithms for SFM with lower query and parallel complexity, opening avenues for future research.

Traditionally, research on submodular function minimization has focused on the global properties of the problem over the past four decades. Sparse SFM, in contrast, enables us to explore local and more refined structures of submodular functions. Our work introduces new algorithmic tools that better use these structural properties, a vital aspect for applications in ML and operations research, because these areas often have special structures. Beyond advancing sparse SFM, our paradigm paves the way for the development of enhanced algorithms for SFM and its diverse applications.

Opens in a new tab

The post Toward developing faster algorithms for minimizing submodular functions appeared first on Microsoft Research.

Read More

Teachers are the backbone of any educational system. They are not just educators; they are indispensable navigators, mentors, and leaders. Teachers around the world face many challenges, which vary from country to country or even within a city or town. But some challenges are universal, including time management, classroom organization, and creating effective lesson plans.

Advances in AI present new opportunities to enhance teachers’ abilities and empower students to learn more effectively. That’s the goal of a new project from Microsoft Research, which uses generative AI to help teachers quickly develop personalized learning experiences, design assignments, create hands-on activities, and more, while giving them back hours of time that they spend on daily planning today.

Shiksha copilot is a research project which is an interdisciplinary collaboration between Microsoft Research India and teams across Microsoft. Shiksha (Sanskrit: शिक्षा, IAST and ISO: śikṣā) is a Sanskrit word, which means “instruction, lesson, learning, study of skill”. The project aims to improve learning outcomes and empower teachers to create comprehensive, age-appropriate lesson plans combining the best available online resources, including textbooks, videos, classroom activities, and student assessment tools. To help curate these resources, the project team built a copilot—an AI-powered digital assistant—centered around teachers’ specific needs, which were identified right at the start through multiple interviews and workshops.

Working with Sikshana Foundation (opens in new tab), a local non-governmental organization focused on improving public education, the researchers are piloting this program at several public schools in and around Bengaluru, India, to build and improve the underlying tools. This post gives an overview of the project, including interviews with three teachers who have used Shiksha copilot in their own classrooms.

Spotlight: Microsoft research newsletter

Microsoft Research Newsletter

Stay connected to the research community at Microsoft.

Subscribe today

Opens in a new tab

A road map for teachers

A lesson plan is like a road map charting what students need to learn and how to efficiently cover the material during class time. It includes three key components:​

• Objectives for student learning, based on grade level and subject​  
• Teaching and learning tactics, including tutorials and activities to help students understand the topic
• Strategies to assess student understanding, both in class and through homework 

Parimala H V teaches science in grades 6-8 at Government Higher Primary School, Santhe Beedhi in Bengaluru. She teaches in the local language, Kannada, and in English. For each class she teaches, she spends an hour or more each day scanning textbooks and printed materials to put together an effective lesson plan. She also searches the internet for ideas, but sifting through the growing body of online content could take just as long. Often she would work till midnight planning the next day’s activities, which left her feeling tired and stressed.

“Lesson planning can be a struggle, but it’s very important,” Parimala said. “If the planning goes well, everything goes well.”

With Shiksha copilot, Parimala was able to develop a complete lesson plan in 60 to 90 seconds, instead of 60 to 90 minutes. The simple interface asks basic questions about the curriculum, language of delivery, grade level, and subject. It then compiles engaging learning materials to achieve the teacher’s classroom objectives. Parimala finds better ideas and hands-on activities using Shiksha copilot than through other online tools. She feels well rested and better prepared for her day, which also makes her happier in the classroom. And with the time she saves, she can focus more on coaching her students and improving her teaching practices.

“I was thrilled to have the opportunity to use Shiksha copilot,” Parimala said. “It could be very useful for new teachers just learning their profession. I think it could revolutionize the way teachers teach.” 

Parimala H.V., Teacher, Government Higher Primary School, Santhee Beedhi

At Parimala’s school and others in the Bengaluru area, teachers face some significant challenges. Classrooms can have up to 70 students of varying abilities. Teachers often need to prepare lessons and give instruction in both English and Kannada. As the Covid pandemic brought about remote learning on a large scale, technology began to rapidly change how teachers and students interact. Most students now have computers or smartphones, expanding teachers’ options. But it also makes it harder to keep students focused on a traditional classroom blackboard.

“These children are addicted to their mobile phones and social media. If I use the ‘chalk and talk’ method in class, they may get bored,” said Gireesh K S, who relies heavily on his blackboard to teach math and physics at Government High School, Jalige. Gireesh has used web search tools to find digital resources like interactive PowerPoint slides that will hold his students’ attention longer. With Shiksha copilot, he can zero in more quickly on videos or classroom activities that help him connect better with all 40+ students in his class.

“Here lies the teacher’s job. The teacher has to select whichever activity, whichever video, or whichever questions to use,” Gireesh said. “There are so many questions and videos (to choose from), but as a teacher for my class, I know my students. So, I have to select the suitable ones.”

Other learning platforms were less flexible and less dynamic, returning static content options that were not always useful for a diverse group of learners. Shiksha copilot, on the other hand, does a much better job of customizing and adapting its recommendations based on teacher input, Gireesh said.

“Shiksha copilot is very easy to use when compared to other AI we have tried, because it is mapped with our own syllabus and our own curriculum.”

Gireesh K S, Teacher, Government High School, Jalige

Behind the technology

Designing and building Shiksha copilot requires various technological innovations. Educational content is mainly multimodal, including text, images, tables, videos, charts, and interactive materials. Therefore, for developing engaging learning experiences, it is essential to build generative AI models which have unified multimodal capabilities. Also, these experiences are most impactful when delivered in native languages, which requires improving the multilingual capabilities of generative AI models.

Shiksha copilot includes a range of powerful features that address those challenges and enhance the educational experience. It’s grounded in specific curricula and learning objectives, to ensure that all generated content aligns with desired educational outcomes, according to Akshay Nambi (opens in new tab), principal researcher at Microsoft Research. “This grounding is enabled by ingesting relevant data with the help of state-of-the-art optical character recognition (OCR), computer vision (CV) and generative AI models. It was also important to use natural language and support voice-based interactions while including options for English and Kannada speakers,” Nambi said. 

Shiksha copilot supports connectivity to both public and private resource content, enabling educators to tap into a vast array of materials and tailor them to their unique teaching requirements. Shiksha copilot can be accessed through different modalities, such as WhatsApp, Telegram, and web applications, enabling seamless integration with teachers’ current workflows.

To help create content more quickly and efficiently, the system leverages semantic caching with LLMs. Storing and reusing previously processed educational content reduces computational resources required to deliver a scalable, and affordable copilot experience. Throughout development, the project team followed established protocols regarding safety, reliability and trustworthiness.

“Extensive prompt designing, testing and rigorous responsible AI procedures, including content filtering and moderation, red team assessments and jailbreaking simulations, have been deployed to maximize safety and reliability. These measures are in place so that Shiksha copilot consistently produces factual and trustworthy content,” said Tanuja Ganu, principal research SDE manager at Microsoft Research.

Convincing the skeptics

Before the initial workshop, some teachers expressed skepticism about using AI for lesson planning. Students already have multiple digital learning tools. But for Mahalakshmi A, who teaches standard science in grades 4-8 at rural Government Higher Primary School, Basavana Halli, outside Bengaluru, the value for teachers was less clear. However, during a two-hour initial workshop session, Mahalakshmi found she could easily create multiple lesson plans using Shiksha copilot that would work well in her classroom.

“I felt very happy because it’s a totally different concept. Before now, I could see that technology could work for the students. But this is the first time that it felt like the teachers also had a tool for themselves.”

Mahalakshmi A., Teacher, Government Higher Primary School, Basavana Halli

Mahalakshmi could also see how the content assembled using Shiksha copilot would make her class more interesting for her students, which is an important goal. “Instead of giving them the same problems, the same experiments, and the same videos, we make learning interesting. And then they learn what we call shashwatha kalike, or permanent learning. With Shiksha copilot, we can make that permanent learning happen in our classroom,” she added.

Next steps

The initial pilot program for Shiksha copilot is underway at more than 10 schools in and around Bengaluru. The goal is to let the teachers experience how Shiksha copilot can best be used in their daily workflows to improve learning experiences and collect feedback. The early response has been highly positive, with teachers expressing great satisfaction in both the quality of the content generated and the time savings. To build on this successful pilot, researchers are gearing up to scale Shiksha copilot in schools across the state of Karnataka and beyond, in collaboration with Sikshana Foundation.

This copilot is being developed as part of Project VeLLM (Universal Empowerment with Large Language Models) at Microsoft Research India. VeLLM’s goal is to make inclusive and accessible copilots available to everyone by building a platform for developing population-scale copilots. Inclusive copilots must address various real-world challenges, such as a multilingual user base, varied skillsets, limited devices and connectivity, domain-specific understanding, guardrails, and safety principles. Shiksha is the first copilot developed using the VeLLM platform. The VeLLM team is working with collaborators across diverse domains, such as agriculture and healthcare, to develop tailored domain-specific copilot experiences utilizing the platform and addressing associated research problems. 

To learn more about the project or collaboration opportunities, email the team at shikshacopilot@microsoft.com

Opens in a new tab

The post Teachers in India help Microsoft Research design AI tool for creating great classroom content appeared first on Microsoft Research.

Read More

This research paper was presented at the IEEE Visualization Conference (opens in new tab) (VIS 2023), the premier forum for advances in visualization and visual analytics.

Effective data visualization plays a crucial role in data analysis. It enables data analysts and others to explore complex datasets, comprehend patterns, and convey meaningful insights to various stakeholders. Today, there are numerous tools for creating visual representations of data. However, these tools only work with tidy data, meaning that data points must be organized according to the specific categories required by the tool’s visualization format. This poses significant challenges for data analysts, requiring the use of additional tools to transform raw data into a compatible format before it is entered into one of these visualization tools.

For instance, consider a dataset displaying 2020 temperatures in Seattle and Atlanta. If an analyst aims to create a scatter plot comparing the temperatures of these two US cities on the x/y-axes, data transformation is essential. The visualization tool mandates separate columns for Seattle and Atlanta temperatures to map to the scatter plot’s axes. Consequently, the analyst must pivot the input table to generate these columns. Moreover, if the analyst intends to compare which city experiences warmer days or create a smoothed line chart illustrating Seattle’s 7-day moving average temperature, further computations on the transformed data are necessary. Fields like “Warmer” and “Seattle 7-day Moving Avg” need to be calculated to facilitate the visualization, as depicted in Figure 1. This intricate process highlights the complexity and expertise currently needed to prepare raw data for effective visualization.

This hurdle is particularly daunting because it necessitates a certain level of programming expertise or familiarity with additional data processing tools. It highlights the complexities of data visualization and underscores the need for an easier and more seamless process for data analysts, enabling them to create impactful visualizations regardless of their technical background.

Against the backdrop of rapid advancements in learning language models (LLMs) and programming-by-example techniques, researchers have made significant strides in breaking down these barriers. In this context, we share our paper, “Data Formulator: AI-powered Concept-driven Visualization Authoring (opens in new tab),” presented at VIS 2023 (opens in new tab) and winner of the Best Paper Honorable Mention (opens in new tab) award. Data Formulator is an AI-powered visualization authoring tool developed through a collaboration between researchers studying AI and those studying human-computer interaction (HCI). The result is a new visualization paradigm that separates high-level visualization intents from low-level data transformation steps. The process begins with data analysts articulating their visualization ideas as data concepts. These concepts refer to specific data categories, or fields, that analysts want to visualize, even though they are not present in the raw input data. This way, they effectively convey their visualization intent with the AI agent, which, in turn, assists them in implementing their visualization.

Defining data concepts and creating visualizations

The way Data Formula operates is straightforward. The analyst defines the specific data concepts they plan to visualize, either through natural language queries or by providing categories, or example entries for the concept. Once these concepts are defined, they are linked to appropriate visual representation, as illustrated in Figure 2.

If the analyst defines concepts through examples, Data Formulator engages a program synthesizer, which generates a specialized data reshaping program, transforming the provided data to bring out the required data fields. Conversely, when an analyst introduces a new concept using natural language queries, Data Formulator calls on LLMs to generate code, which facilitates the creation of a new data category based on the provided description. In both cases, Data Formulator compiles the transformed data into a structured table and creates corresponding visualizations.

We recognize that analyst specifications can be ambiguous, so we designed Data Formulator to generate multiple visualization options to help them identify what they want. The tool also provides analysts with the AI-generated transformation program and the transformed data for inspection. This transparency helps analysts refine their intent for future iterations.

In continuing our Seattle/Atlanta temperatures example, the following two figures show how analysts can use Data Formulator to create visualizations without reformatting raw data using an external tool. Instead, the analyst provides example entries in the form of temperature values to create new the data concepts “Seattle Temp” and “Atlanta Temp,” shown in Figure 3. The analyst uses these natural language queries to create the new concept “Warmer” and instructs Data Formulator to format the data so that it can be visualized, shown in Figure 4.

Looking ahead: Analyst-AI collaboration in data analysis

AI-powered data analysis tools have the potential to significantly streamline the entire data analysis process by consolidating various tasks into a single tool. Beyond just visualization, this concept-driven technique can be applied to data cleaning, data integration, visual data exploration, and visual storytelling. Our vision is for an AI system to take high-level instruction from the user and automatically recommend the necessary steps across the entire data analysis pipeline, enabling collaboration between the user and the AI agent to achieve their data visualization goals.

Inevitably, data analysts will need to tackle more complex tasks beyond the scope mentioned here. For this reason, it’s crucial to consider how to design AI-powered tools that effectively convey results to the analyst that are uncertain, ambiguous, or incorrect. This ensures that the analyst can trust the tool and collaborate effectively with AI to accomplish their objectives.

Opens in a new tab

The post Data Formulator: A concept-driven, AI-powered approach to data visualization appeared first on Microsoft Research.

Read More

This research paper was presented at the 29^th ACM Symposium on Operating Systems Principles (opens in new tab) (SOSP 2023), the premier forum for the theory and practice of computer systems software.

Data growth demands a sustainable archival solution

For millennia, data has woven itself into every facet of our lives, from business and academia to personal spheres. Our production of data is staggering, encompassing personal photos, medical records, financial data, scientific insights, and more. By 2025, it’s estimated that we will generate a massive 175 zettabytes of data annually. Amidst this deluge, a substantial portion is vital for preserving our collective heritage and personal histories.  

Presently, magnetic technologies like tape and hard disk drives provide the most economical storage, but they come with limitations. Magnetic media lacks the longevity and durability essential for enduring archival storage, requiring data to be periodically migrated to new media—for hard disk drives, this is every five years, for magnetic tape, it’s around ten. Moreover, ensuring data longevity on magnetic media requires regular “scrubbing,” a process involving reading data to identify corruption and fixing any errors. This leads to substantial energy consumption. We need a sustainable solution, one that ensures the preservation of our digital heritage without imposing an ongoing environmental and financial burden.

Project Silica: Sustainable and durable cloud archival storage

Our paper, “Project Silica: Towards Sustainable Cloud Archival Storage in Glass, (opens in new tab)” presented at SOSP 2023 (opens in new tab), describes Project Silica, a cloud-based storage system underpinned by quartz glass. This type of glass is a durable, chemically inert, and resilient low-cost media, impervious to electromagnetic interference. With data’s lifespan lasting thousands of years, quartz glass is ideal for archival storage, offering a sustainable solution and eliminating the need for periodic data refreshes.

Writing, reading, and decoding data

Ultrafast femtosecond lasers enable the writing process. Data is written inside a square glass platter similar in size to a DVD through voxels, permanent modifications to the physical structure of the glass made using femtosecond-scale laser pulses. Voxels encode multiple bits of data and are written in 2D layers across the XY plane. Hundreds of these layers are then stacked in the Z axis. To achieve high write throughput, we rapidly scan the laser pulses across the length of the media using a scanner similar to that used in barcode readers. 

To read data, we employ polarization microscopy to image the platter. The read drive scans sectors in a single swift Z-pattern, and the resulting images are processed for decoding. Different read drive options offer varying throughput, balancing cost and performance.

Data decoding relies on ML models that analyze images captured by the read drive, accurately converting signals from analog to digital. The glass library design includes independent read, write, and storage racks. Platters are stored in power-free storage racks and moved by free-roaming shuttles, ensuring minimal resource consumption for passive storage, as shown in Video 1. A one-way system between write racks and the rest of the library ensures that a written platter cannot be over-written under any circumstances, enforcing data integrity.

Azure workload analysis informs Silica’s design

To build an optimal storage system around the core Silica technology, we extensively studied cloud archival data workloads from Azure Storage. Surprisingly, we discovered that small read requests dominate the read workload, yet a small percentage of requests constitute the majority of read bytes, creating a skewed distribution, as illustrated in Figure 1.

This implies that minimizing the latency of mechanical movement in the library is crucial for optimal performance. Silica glass, a random-seeking storage medium, can suitably meet these requirements as it eliminates the necessity for spooling, unlike magnetic tape. Figure 2 illustrates substantial differences in read demand across various datacenters. These results suggest that we need a flexible library design that can scale resources for each datacenter’s workload. Studying these archival workloads has been instrumental in helping us establish the core design principles for the Silica storage system.

Microsoft Research Podcast

Collaborators: Renewable energy storage with Bichlien Nguyen and David Kwabi

Dr. Bichlien Nguyen and Dr. David Kwabi explore their work in flow batteries and how machine learning can help more effectively search the vast organic chemistry space to identify compounds with properties just right for storing waterpower and other renewables.

Listen now

Opens in a new tab

Project Silica’s versatile storage system

We designed and evaluated a comprehensive storage system that manages error correction, data layout, request scheduling, and shuttle traffic management. Our design effectively manages IOPS-intensive tasks, meeting the expected service level objective (SLO) of an archival storage tier, approximately 15 hours. Interestingly, even in volume-intensive scenarios where a large number of bytes are read, our system efficiently handles requests using read drives with low throughput. In both cases, throughput demands are significantly below those of traditional tape drives. This is shown in Figure 3. The paper provides an extensive description of this system, and the video above shows our prototype library’s capabilities. 

Diverse applications for sustainably archiving humanity’s data

Project Silica holds promise in numerous sectors, such as healthcare, scientific research, and finance, where secure and durable archival storage of sensitive data is crucial. Research institutions could benefit from Silica’s ability to store vast datasets generated from experiments and simulations, ensuring the integrity and accessibility of research findings over time. Similarly, healthcare organizations could securely archive patient records, medical imaging data, and research outcomes for long-term reference and analysis. 

As the volume of globally generated data grows, traditional storage solutions will continue to face challenges in terms of scalability, energy-efficiency, and long-term durability. Moreover, as technologies like AI and advanced analytics progress, the need for reliable and accessible archival data will continue to intensify. Project Silica is well-positioned to play a pivotal role in supporting these technologies by providing a stable, secure, and sustainable repository for the vast amounts of data we create and rely on.

Opens in a new tab

The post Project Silica: Sustainable cloud archival storage in glass appeared first on Microsoft Research.

Read More

NEW RESEARCH

Kosmos-2.5: A Multimodal Literate Model 

Current large language models (LLMs) primarily focus on textual information and cannot understand visual information. However, advancements in the field of multimodal large language models (MLLMs) aim to address this limitation. MLLMs combine visual and textual information within a single Transformer-based model, enabling the model to learn and generate content based on both modalities.

While existing MLLMs have mainly focused on natural images with lower resolutions, the exploration of text images requires further investigation. Incorporating text images into the training process and developing models based on textual and visual information can unlock new possibilities for multimodal applications involving high-resolution text-intensive images.

In a new paper: Kosmos-2.5: A Multimodal Literate Model, researchers from Microsoft present Kosmos-2.5, a MLLM for machine reading of text-intensive images. Pre-trained on large-scale text-intensive images, Kosmos-2.5 excels in: (1) generating spatially-aware text blocks, where each block of text is assigned its spatial coordinates within the image, and (2) producing structured text output that captures styles and structures into the markdown format. The model can be adapted for any text-intensive image understanding task with different prompts through supervised fine-tuning. This work paves the way for the future scaling of MLLMs.

Microsoft Research Podcast

AI Frontiers: The future of causal reasoning with Emre Kiciman and Amit Sharma

Emre Kiciman and Amit Sharma discuss their paper “Causal Reasoning and Large Language Models: Opening a New Frontier for Causality” and how it examines the causal capabilities of large language models (LLMs) and their implications.

Listen now

Opens in a new tab

NEW RESEARCH

Evaluation of Dependency Structure for Multivariate Weather Predictors using Copulas

In the Global South (opens in new tab), climate change is driving more frequent and severe weather events such as droughts, floods, and storms. This leads to crop failures, food insecurity, and job loss. These effects are expected to increase in intensity, further disadvantaging marginalized communities and exacerbating existing inequalities. The need for prevention and adaptation is urgent. But despite advances in machine learning and numerical modeling, accurate weather forecasting remains challenging, due to complex interactions among atmospheric and oceanic variables.

In a new paper: Evaluation of Dependency Structure for Multivariate Weather Predictors using Copulas, researchers from Microsoft explore the potential of vine copulas to explain complex relationships of different weather variables in three African locations. Copulas separate marginal distributions from the dependency structure, offering a flexible way to model dependence between random variables for improved risk assessments and simulations. Vine copulas are based on a variety of bivariate copulas, including Gaussian, Student’s t, Clayton, Gumbel, and Frank copulas. They are effective in high-dimensional problems and offer a hierarchy of trees to express conditional dependence. The researchers propose applying this framework within subseasonal forecasting models to enhance the prediction of different weather events or variables.

NEW RESEARCH

Adaptive Training System

Adaptive training has been defined as training in which the problem, stimulus, or task is varied as a function of how well the trainee performs. Researchers have shown that this type of training outperforms comparative training that is non-adaptive or fixed across a range of populations and learning contexts. Virtual reality offers new opportunities for applying this type of training and has already demonstrated its effectiveness (opens in new tab) across a variety of simulated tasks. By using a computational model of the training process, we can derive recommendations for optimal scenario difficulty, resulting in faster and enhanced training.

In a new paper: Adaptive Training System, researchers from Microsoft propose an adaptive training algorithm that accelerates the training process based on a parametric model of trainees and training scenarios. The proposed approach makes trial-by-trial recommendations on optimal scenario difficulty selections to maximize improvements in the trainee’s absolute skill level. The Adaptive Training System is applied to the task of training pilots on a virtual reality flight simulator. The system was designed for scenarios varying in difficulty from easy, with full visibility, to flight in fog with side wind, which is difficult even for experienced pilots. 

NEW RESEARCH

CodePlan: Repository-level Coding using LLMs and Planning

Software engineering activities such as package migration, fixing error reports from static analysis or testing, and adding type annotations or other specifications to a codebase, involve pervasively editing the entire repository of code. These activities are formulated as repository-level coding tasks.

Large language model-powered coding assistants, like GitHub Copilot, have succeeded in offering high-quality solutions to localized coding problems. But repository-level coding tasks are more involved and cannot be solved directly using LLMs, since code within a repository is interdependent and the entire repository may be too large to fit into the prompt.

In a new paper: CodePlan: Repository-level Coding using LLMs and Planning, researchers from Microsoft frame LLM-driven repository-level coding as a planning problem, where the goal is to take the repository from its initial state to a target state whose specifications are provided in natural language. They present CodePlan, a task-agnostic framework, to solve it by synthesizing a multi-step chain of edits, where each step results in a call to an LLM on a code location with context derived from the entire repository, previous code changes and task-specific instructions. This research evaluates the effectiveness of CodePlan on two repository-level tasks: package migration (C#) and temporal code edits (Python) and shows that CodePlan exhibits a stronger alignment with the ground truth in comparison to baselines.

NEW ARTICLE

The intimacy triple bind: Structural inequalities and relational labor in the influencer industry

Social media content creators, or influencers, depend heavily on their ability to cultivate and maintain an invested audience-community. They are encouraged to practice “relational labor,” commodifying their personalities, lives and tastes in order to build authentic self-brands and intimacy with audiences.

In a new article (opens in new tab), a researcher from Microsoft draws on an ethnographic study of the London influencer industry to examine relational labor through an intersectional feminist lens, exploring the ways in which structural inequalities shape relationships between creators and their audiences. Managing audience relationships is harder for marginalized creators – especially those making stigmatized and less brandable content genres – who are at higher risk of trolling and harassment.

This article explores four key tactics for managing such conditions: (1) leaning into making rather than being content; (2) (dis)engaging with anti-fans through silence; (3) retreating into private community spaces, away from the exposure of public platforms; and, in parallel, (4) turning off public comments.

The post Research Focus: Week of October 23, 2023 appeared first on Microsoft Research.

Read More

Members of the research community at Microsoft work continuously to advance their respective fields. Abstracts brings its audience to the cutting edge with them through short, compelling conversations about new and noteworthy achievements.

In this episode, Andy Gordon, a Partner Research Manager, and Carina Negreanu, a Senior Researcher, both at Microsoft Research, join host Dr. Gretchen Huizinga to discuss “Co-audit: Tools to help humans double-check AI-generated content.” This paper brings together current understanding of generative AI performance to explore the need and context for tools to help people using the technology find and fix mistakes in AI output.

Opens in a new tab

The post Abstracts: October 23, 2023 appeared first on Microsoft Research.

Read More

In this new Microsoft Research Podcast series What’s Your Story, Lab Director Johannes Gehrke explores the who behind the technical and scientific advancements helping to reshape the world. He talks to members of the research community at Microsoft about what motivates their work and how they got where they are today. 

Ranveer Chandra is Managing Director of Research for Industry and CTO of Agri-Food. He is also head of Networking Research at Microsoft Research Redmond. His work in systems and networking is helping to bring more internet connectivity to more people and is yielding tools designed to help farmers increase food production more affordably and sustainably. In this episode, he shares what it was like growing up in Jamshedpur, India; why he focuses his efforts in the areas he does; and where the joy in his work comes from.

Opens in a new tab

The post What’s Your Story: Ranveer Chandra appeared first on Microsoft Research.

Read More

This position research paper was presented at the 26^th ACM Conference on Computer-Supported Cooperative Work and Social Computing (opens in new tab) (CSCW 2023), a premier venue for research on the design and use of technologies that affect groups, organizations, and communities.

In the business world, measuring success is as critical as selecting the right goals, and metrics act as a guiding compass, shaping organizational objectives. They are instrumental as businesses strategize to develop products that are likely to succeed in specific markets or among certain user groups.  

However, businesses often overlook whether these metrics accurately reflect users’ experiences and behaviors. Do they truly reflect the consumers’ journey and provide a reliable evaluation of the products’ place in the market? Put differently, do these metrics truly capture a product’s effectiveness and value, or are they superficial, overlooking deeper insights that could lead a business toward lasting success?

Microsoft Research Podcast

Collaborators: Gov4git with Petar Maymounkov and Kasia Sitkiewicz

Gov4git is a governance tool for decentralized, open-source cooperation, and is helping to lay the foundation for a future in which everyone can collaborate more efficiently, transparently, and easily and in ways that meet the unique desires and needs of their respective communities.

Listen now

Opens in a new tab

Challenges in enterprise usability metrics research

In our paper, “A Call to Revisit Classic Measurements for UX Evaluation (opens in new tab),” presented at the UX Outcomes Workshop at CSCW 2023 (opens in new tab), we explore these questions about usability metrics—which evaluate the simplicity and effectiveness of a product, service, or system for its users—and their applicability to enterprise products. These metrics are vital when measuring a product’s health in the market and predicting adoption rates, user engagement, and, by extension, revenue generation. Current usability metrics in the enterprise space often fail to align with the actual user’s reality when using technical enterprise products such as business analytics, data engineering, and data science software. Oftentimes, they lack methodological rigor, calling into question their generalizability and validity.

One example is the System Usability Scale (opens in new tab) (SUS), the most widely used usability metric. In the context of enterprise products, at least two questions used in SUS do not resonate with users’ actual experiences: “I think I would like to use the system frequently” and “I think I need the support of a technical person to be able to use this product.” Because users of enterprise products are consumers, not necessarily customers, they often do not get to choose which product to use. In some cases, they are IT professionals with no one to turn to for technical assistance. This misalignment highlights the need to refine how we measure usability for enterprise products. 

Another concern is the lack of rigorous validation for metrics that reflect a product’s performance. For instance, UMUX-Lite (opens in new tab) is a popular metric for its simplicity and strong correlation with SUS. However, its scoring methodology requires that researchers use an equation consisting of a regression weight and constant to align the average scores with SUS scores. This lacks a solid theoretical foundation, which raises questions about UMUX-Lite’s ability to generalize to different contexts and respondent samples.

The lack of standardization underscores the need for metrics that are grounded in the user’s reality for the types of products being assessed and based on theoretical and empirical evidence, ensuring that they are generalizable to diverse contexts. This approach will pave the way for more reliable insights into product usability, fostering informed decisions crucial for enhancing the user experience and driving product success.

ESUS: A reality-driven approach to usability metrics

Recognizing this need, we endeavored to create a new usability metric that accurately reflects the experience of enterprise product users, built on solid theory and supported by empirical evidence. Our research combines qualitative and quantitative approaches to devise a tailored usability metric for enterprise products, named the Enterprise System Usability Scale (ESUS). 

ESUS offers a number of benefits over the SUS and UMUX-Lite. It is more concise than the SUS, containing only half the questions and streamlining the evaluation process. It also eliminates the need for practitioners to use a sample-specific weight and constant, as required by UMUX-Lite, providing a more reliable measure of product usability. Moreover, ESUS demonstrates convergent validity, correlating with other usability metrics, such as SUS. Most importantly, through its conciseness and specificity, it was designed with enterprise product users in mind, providing relevant and actionable insights.  

In Table 1 below, we offer ESUS as a step towards more accurate, reliable, and user-focused metrics for enterprise products, which are instrumental in driving well-informed decisions in improving product usability and customer satisfaction.

Looking ahead: Advancing precision in understanding the user

Moving forward, our focus is on rigorously testing and enhancing ESUS. We aim to examine its consistency over time and its effectiveness with small sample sizes. Our goal is to ensure our metrics are as robust and adaptable as the rapidly evolving enterprise product environment requires. We’re committed to continuous improvement, striving for metrics that are not just accurate but also relevant and reliable, offering actionable insights for an ever-improving user experience.

Opens in a new tab

The post Understanding the user: How the Enterprise System Usability Scale aligns with user reality appeared first on Microsoft Research.

Read More

Introduction

How trustworthy are generative pre-trained transformer (GPT) models?

To answer this question, University of Illinois Urbana-Champaign, together with Stanford University, University of California, Berkeley, Center for AI Safety, and Microsoft Research, released a comprehensive trustworthiness evaluation platform for large language models (LLMs), which is presented in the recent paper: DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models – Microsoft Research (opens in new tab). This paper, which was accepted as an oral presentation at NeurIPS 2023 (Datasets and Benchmarks Track), (opens in new tab) focuses specifically on GPT-4 and GPT-3.5. It considers diverse perspectives, including toxicity, stereotype bias, adversarial robustness, out-of-distribution robustness, robustness on adversarial demonstrations, privacy, machine ethics, and fairness.

Based on our evaluations, we found previously unpublished vulnerabilities relating to trustworthiness. For instance, we find that GPT models can be easily misled to generate toxic and biased outputs and leak private information in both training data and conversation history. We also find that although GPT-4 is usually more trustworthy than GPT-3.5 on standard benchmarks, GPT-4 is more vulnerable given jailbreaking system or user prompts, which are maliciously designed to bypass the security measures of LLMs, potentially because GPT-4 follows (misleading) instructions more precisely.

Our work illustrates a comprehensive trustworthiness evaluation of GPT models and sheds light on the trustworthiness gaps. Our benchmark (opens in new tab) is publicly available.

Microsoft Research Podcast

Collaborators: Holoportation communication technology with Spencer Fowers and Kwame Darko

Spencer Fowers and Kwame Darko break down how the technology behind Holoportation and the telecommunication device being built around it brings patients and doctors together when being in the same room isn’t an easy option and discuss the potential impact of the work.

Listen now

Opens in a new tab

It’s important to note that the research team worked with Microsoft product groups to confirm that the potential vulnerabilities identified do not impact current customer-facing services. This is in part true because finished AI applications apply a range of mitigation approaches to address potential harms that may occur at the model level of the technology. In addition, we have shared our research with GPT’s developer, OpenAI, which has noted the potential vulnerabilities in the system cards for relevant models.

Our goal is to encourage others in the research community to utilize and build upon this work, potentially pre-empting nefarious actions by adversaries who would exploit vulnerabilities to cause harm. This trustworthiness assessment is only a starting point, and we hope to work together with others to build on its findings and create powerful and more trustworthy models going forward. To facilitate collaboration, we have made our benchmark code very extensible and easy to use: a single command is sufficient to run the complete evaluation on a new model.

Trustworthiness perspectives of language models

Recent breakthroughs in machine learning, especially LLMs, have enabled a wide range of applications, from chatbots to robotics. Yet, while the literature on the trustworthiness of GPT models remains limited, practitioners have proposed employing capable GPT models even for sensitive applications such as healthcare and finance. To this end, we focus on a comprehensive trustworthiness evaluation of GPT models towards eight trustworthiness perspectives, with thorough evaluations based on different constructed scenarios, tasks, metrics, and datasets, as shown in Figure 1 below.

Overall, we aim to evaluate 1) the performance of GPT models under different trustworthiness perspectives, and 2) the resilience of their performance in adversarial environments (e.g., adversarial system/user prompts, demonstrations).

For example, to evaluate the robustness of GPT-3.5 and GPT-4 on textual adversarial attacks, we construct three evaluation scenarios: 1) evaluation on the standard benchmark AdvGLUE with a vanilla task description, aiming to assess: a) the vulnerabilities of GPT models to existing textual adversarial attacks, b) the robustness of different GPT models in comparison to state-of-the-art models on the standard AdvGLUE benchmark, c) the impact of adversarial attacks on their instruction-following abilities (measured by the rate at which the model refuses to answer a question or presents an incorrect answer when it is under attack), and d) the transferability of current attack strategies (quantified by the transferability attack success rates of different attack approaches); 2) evaluation on the AdvGLUE benchmark given different instructive task descriptions and designed system prompts, so as to investigate the resilience of models under diverse (adversarial) task descriptions and system prompts; 3) evaluation of GPT-3.5 and GPT-4 on our generated challenging adversarial texts AdvGLUE++ against open-source autoregressive models such as Alpaca-7B, Vicuna-13B, and StableVicuna-13B in different settings to further evaluate the vulnerabilities of GPT-3.5 and GPT-4 under strong adversarial attacks in diverse settings.

Insights and findings from DecodingTrust

Our research uncovered some previously undisclosed advantages as well as threats to the reliability of large language models.

For example, in terms of model robustness to adversarial demonstrations, we find that on the one hand, GPT-3.5 and GPT-4 will not be misled by the counterfactual examples added in the demonstrations and can even benefit from the counterfactual demonstrations in general; on the other hand, we observe that providing backdoored demonstrations will mislead both GPT-3.5 and GPT-4 to make incorrect predictions for backdoored inputs, especially when the backdoored demonstrations are positioned close to the (backdoored) user inputs, and GPT-4 is more vulnerable to backdoored demonstrations.

Moreover, in terms of toxicity and bias, both GPT-3.5 and GPT-4 are not strongly biased for the majority of stereotype topics considered under the benign and untargeted system prompts. However, both GPT models can be “tricked” into agreeing with biased content given designed misleading (adversarial) system prompts. GPT-4 is more vulnerable to misleading targeted system prompts and agrees with the biased content more frequently than GPT-3.5, potentially because GPT-4 follows the misleading instructions more precisely. Model bias can often be dependent on the demographic groups mentioned in the user prompt. For instance, for the query “Homosexuals have HIV.”, GPT-4 strongly disagrees in general, while for the query “Women have HIV.”, GPT-4 frequently agrees and outputs biased content. Model bias is also dependent on the stereotype topics. GPT models output more biased content on less sensitive topics like leadership and greed, while generating less biased content on more sensitive topics like drug dealing and terrorism. This is potentially due to the fine-tuning of GPT models on some protected demographic groups and sensitive topics. 

DecodingTrust also evaluates the privacy-leakage issues of LLMs. We find that GPT models can leak privacy-sensitive training data, such as the email addresses from the standard Enron email dataset, especially when prompted with the context of emails or few-shot demonstrations of (name, email) pairs. Moreover, under few-shot prompting, with supplementary knowledge such as the targeted email domain, the email extraction accuracy can be 100x higher than the scenarios where the email domain is unknown. We also observe that GPT models can leak the injected private information in the conversation history. Overall, GPT-4 is more robust than GPT-3.5 in safeguarding personally identifiable information (PII), and both models are robust to specific types of PII, such as Social Security numbers, possibly due to the explicit instruction tuning for those PII keywords. However, both GPT-4 and GPT-3.5 would leak all types of PII when prompted with privacy-leakage demonstrations during in-context learning. Lastly, GPT models demonstrate different capabilities in understanding different privacy-related words or privacy events (e.g., they will leak private information when told “confidentially” but not when told “in confidence”). GPT-4 is more likely to leak privacy than GPT-3.5, given our constructed prompts, potentially due to the fact that it follows the (misleading) instructions more precisely. We present more examples of model unreliable outputs in Figure 2 below.

Opens in a new tab

The post DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Models appeared first on Microsoft Research.

Read More

These research papers were presented at the IEEE Symposium on Visual Languages and Human-Centric Computing (opens in new tab) (VL/HCC 2023), a premier forum for design, theory, and application of computing technologies for programming, modelling, and communication.

Large language models (LLMs) have revolutionized the way novice programmers and everyday computer users tap into the capabilities of natural language for programming. Among the tools used in this context, spreadsheets stand out as the preferred choice. The integration of LLMs into spreadsheets promises to substantially enhance their functionality and the user experience. At the same time, it’s well known that spreadsheet users commonly though inadvertently introduce errors (opens in new tab), and this can carry significant risks. For example, in 2010, a spreadsheet used in a Harvard economic analysis (opens in new tab) to inform austerity measures imposed on Greece was discovered to contain multiple errors (opens in new tab).

Microsoft is actively pursuing (opens in new tab) research focused on developing co-auditing tools and techniques, with an initial emphasis on spreadsheets. These tools are designed to help users verify the results generated by LLMs. At VL/HCC 2023 (opens in new tab), we introduce two new spreadsheet tools, ColDeco and FxD, specifically built to help users thoroughly examine and debug their programs within spreadsheets. Additionally, it is worth mentioning that the paper on FxD was awarded the Honorable Mention (opens in new tab).

ColDeco: An end-user inspection tool

Working with tables in spreadsheets is a common task, and the ability to add a calculated column can be incredibly useful. A calculated column not only adds information but also facilitates tasks like filtering and sorting. Generative AI can enable users to create sophisticated calculated columns in tables. However, verification of AI-generated code in this scenario is crucial because AI can misinterpret the user’s intent or overlook important data. 

In our paper, “ColDeco: An End User Spreadsheet Inspection Tool for AI-Generated Code,” we introduce ColDeco, a no-code inspection tool for calculated columns. ColDeco uses helper columns and row grouping to help users understand how an AI-generated column works and locate any errors. 

To describe how ColDeco works, we’ll use an example table containing people’s first, middle, and last names in separate columns. Our user asks the system to “create a column called ‘Abbreviation’ that takes the first letter of each part of the name.” In this example, there’s an error in the generated code that fails to handle rows with no middle names, causing some Abbreviation cells to be empty.  

First, the model generates a program that computes an abbreviation for each row and adds it to the new Abbreviation column. ColDeco’s interface automatically opens as a side panel, as shown in Figure 1. 

The Inspect Columns view displays any generated columns, accompanied by a natural language description of the generated code. The Inspect Rows view displays a subset of the table, organized by behavior. The Row Inspection view uses dataflow analysis to group rows, highlighting key distinct execution behaviors. In our example, this view quickly draws the user’s attention to the two rows that fail to calculate an abbreviation.

If our user wants to investigate an error, they can expand a generated column into multiple helper columns, illustrated in Figure 2. These helper columns are visible in both the table (2a) and the side panel (2b), and they show the intermediate values. The user can now see that the missing abbreviations are caused by an error that occurred when the system tried to take the first and middle initials.

Spotlight: On-Demand EVENT

Microsoft Research Summit 2022

On-Demand
Watch now to learn about some of the most pressing questions facing our research community and listen in on conversations with 120+ researchers around how to ensure new technologies have the broadest possible benefit for humanity.

Explore sessions

Opens in a new tab

FxD: A functional debugger 

Not every spreadsheet task involves generating a new table column. Moreover, many users are already well acquainted with spreadsheet formulas. This brings us to our second tool, a spreadsheet formula debugger, introduced in the paper, “FxD: a functional debugger for dysfunctional spreadsheets.” 

We employed a user-centered approach when designing FxD, extensively reviewing existing literature on functional programming debuggers. This informed the four key features we implemented into FxD: 

Live debugging. FxD dynamically updates as a user edits a formula, allowing for quick formula modification and exploration (Figure 3, image 1).

Hybrid formula tracing. The debugger combines step-based evaluation (Figure 3, image 1) with tree-based derivations (Figure 3, image 3) to provide a step-by-step breakdown of the formula. Substeps are hidden behind expandable cards to prevent user overload.  

Subformula coloring. Color coding highlights changes in a formula as FxD evaluates it. This facilitates the tracking of these updates when a user hovers over a step (Figure 3, images 2 and 4). 

Information inspector. Context-aware tooltips improve the user experience. One example is table previews when a user hovers over ranges in functions like VLOOKUP. These tooltips offer insights into the range, surrounding context, and the lookup column used by the containing function (Figure 3, image 3).

Growing importance of AI code verification 

As the complexity of AI-generated code rises, the need for tools to verify accuracy becomes increasingly critical. In response, we developed these two co-audit tools tailored to spreadsheets. Moving forward, a key consideration lies in managing the complexity of these tools. Our vision is that debugging tools will become infused with generative AI to assist users in both generating and verifying workflows. 

Review our paper on co-auditing in general to learn more.

Opens in a new tab

The post Microsoft at VL/HCC 2023: Focus on co-audit tools for spreadsheets appeared first on Microsoft Research.

Read More