AI for Productivity Tools
Abstract
This paper aims to explore fundamental questions in the era when AI coding assistants like GitHub Copilot are widely adopted: what do developers truly value and criticize in AI coding assistants, and what does this reveal about their needs and expectations in real-world software development? Unlike previous studies that conduct observational research in controlled and simulated environments, we analyze extensive, first-hand user reviews of AI coding assistants, which capture developers' authentic perspectives and experiences drawn directly from their actual day-to-day work contexts. We identify 1,085 AI coding assistants from the Visual Studio Code Marketplace. Although they only account for 1.64% of all extensions, we observe a surge in these assistants: over 90% of them are released within the past two years. We then manually analyze the user reviews sampled from 32 AI coding assistants that have sufficient installations and reviews to construct a comprehensive taxonomy of user concerns and feedback about these assistants. We manually annotate each review's attitude when mentioning certain aspects of coding assistants, yielding nuanced insights into user satisfaction and dissatisfaction regarding specific features, concerns, and overall tool performance. Built on top of the findings-including how users demand not just intelligent suggestions but also context-aware, customizable, and resource-efficient interactions-we propose five practical implications and suggestions to guide the enhancement of AI coding assistants that satisfy user needs.
This paper aims to explore fundamental questions in the era when AI coding assistants like GitHub Copilot are widely adopted: what do developers truly value and criticize in AI coding assistants, and what does this reveal about their needs and expectations in real-world software development? Unlike previous studies that conduct observational research in controlled and simulated environments, we analyze extensive, first-hand user reviews of AI coding assistants, which capture developers' authentic perspectives and experiences drawn directly from their actual day-to-day work contexts. We identify 1,085 AI coding assistants from the Visual Studio Code Marketplace. Although they only account for 1.64% of all extensions, we observe a surge in these assistants: over 90% of them are released within the past two years. We then manually analyze the user reviews sampled from 32 AI coding assistants that have sufficient installations and reviews to construct a comprehensive taxonomy of user concerns and feedback about these assistants. We manually annotate each review's attitude when mentioning certain aspects of coding assistants, yielding nuanced insights into user satisfaction and dissatisfaction regarding specific features, concerns, and overall tool performance. Built on top of the findings-including how users demand not just intelligent suggestions but also context-aware, customizable, and resource-efficient interactions-we propose five practical implications and suggestions to guide the enhancement of AI coding assistants that satisfy user needs.
Abstract
This study investigates the implementation and efficacy of DeputyDev, an AI-powered code review assistant developed to address inefficiencies in the software development process. The process of code review is highly inefficient for several reasons, such as it being a time-consuming process, inconsistent feedback, and review quality not being at par most of the time. Using our telemetry data, we observed that at TATA 1mg, pull request (PR) processing exhibits significant inefficiencies, with average pick-up and review times of 73 and 82 hours, respectively, resulting in a 6.2 day closure cycle. The review cycle was marked by prolonged iterative communication between the reviewing and submitting parties. Research from the University of California, Irvine indicates that interruptions can lead to an average of 23 minutes of lost focus, critically affecting code quality and timely delivery. To address these challenges, we developed DeputyDev's PR review capabilities by providing automated, contextual code reviews. We conducted a rigorous double-controlled A/B experiment involving over 200 engineers to evaluate DeputyDev's impact on review times. The results demonstrated a statistically significant reduction in both average per PR (23.09%) and average per-line-of-code (40.13%) review durations. After implementing safeguards to exclude outliers, DeputyDev has been effectively rolled out across the entire organisation. Additionally, it has been made available to external companies as a Software-as-a-Service (SaaS) solution, currently supporting the daily work of numerous engineering professionals. This study explores the implementation and effectiveness of AI-assisted code reviews in improving development workflow timelines and code.
This study investigates the implementation and efficacy of DeputyDev, an AI-powered code review assistant developed to address inefficiencies in the software development process. The process of code review is highly inefficient for several reasons, such as it being a time-consuming process, inconsistent feedback, and review quality not being at par most of the time. Using our telemetry data, we observed that at TATA 1mg, pull request (PR) processing exhibits significant inefficiencies, with average pick-up and review times of 73 and 82 hours, respectively, resulting in a 6.2 day closure cycle. The review cycle was marked by prolonged iterative communication between the reviewing and submitting parties. Research from the University of California, Irvine indicates that interruptions can lead to an average of 23 minutes of lost focus, critically affecting code quality and timely delivery. To address these challenges, we developed DeputyDev's PR review capabilities by providing automated, contextual code reviews. We conducted a rigorous double-controlled A/B experiment involving over 200 engineers to evaluate DeputyDev's impact on review times. The results demonstrated a statistically significant reduction in both average per PR (23.09%) and average per-line-of-code (40.13%) review durations. After implementing safeguards to exclude outliers, DeputyDev has been effectively rolled out across the entire organisation. Additionally, it has been made available to external companies as a Software-as-a-Service (SaaS) solution, currently supporting the daily work of numerous engineering professionals. This study explores the implementation and effectiveness of AI-assisted code reviews in improving development workflow timelines and code.
Economics of Productivity
Abstract
As computational capacity increases, it becomes possible to model health systems in greater detail. Multi-disease health system models (HSMs) represent a new development, building on individual level epidemiological models of multiple diseases and capturing how healthcare delivery systems respond to population health needs. The Thanzi la Onse (TLO) model of Malawi is the first of its kind in these respects. In this article, we discuss how we have been bringing economic concepts into the TLO model, and how we are continuing to develop this line of research. This has involved incorporating more sophisticated approaches to account for the effects of the unavailability of healthcare workers, and we are working towards establishing the role of different forms of ownership of healthcare facilities and different management practices. Not only does this broad approach make the model more flexible as a tool for understanding the impact of resource constraints, it opens up the possibility of analysing considerably richer policy scenarios; for example establishing an estimate of the health gain that could be achieved through expanding the workforce or reducing healthcare worker absence.
As computational capacity increases, it becomes possible to model health systems in greater detail. Multi-disease health system models (HSMs) represent a new development, building on individual level epidemiological models of multiple diseases and capturing how healthcare delivery systems respond to population health needs. The Thanzi la Onse (TLO) model of Malawi is the first of its kind in these respects. In this article, we discuss how we have been bringing economic concepts into the TLO model, and how we are continuing to develop this line of research. This has involved incorporating more sophisticated approaches to account for the effects of the unavailability of healthcare workers, and we are working towards establishing the role of different forms of ownership of healthcare facilities and different management practices. Not only does this broad approach make the model more flexible as a tool for understanding the impact of resource constraints, it opens up the possibility of analysing considerably richer policy scenarios; for example establishing an estimate of the health gain that could be achieved through expanding the workforce or reducing healthcare worker absence.
Abstract
This paper investigates the impact of the global financial crisis on the shape of economics as a discipline by analyzing EconLit-indexed journals from 2006 to 2020 using a multilayer network approach. We consider two types of social relationships among journals, based on shared editors (interlocking editorship) and shared authors (interlocking authorship), as well as two forms of intellectual proximity, derived from bibliographic coupling and textual similarity. These four dimensions are integrated using Similarity Network Fusion to produce a unified similarity network from which journal communities are identified. Comparing the field in 2006, 2012, and 2019 reveals a high degree of structural continuity. Our findings suggest that, despite changes in research topics after the crisis, fundamental social and intellectual relationships among journals have remained remarkably stable. Editorial networks, in particular, continue to shape hierarchies and legitimize knowledge production.
This paper investigates the impact of the global financial crisis on the shape of economics as a discipline by analyzing EconLit-indexed journals from 2006 to 2020 using a multilayer network approach. We consider two types of social relationships among journals, based on shared editors (interlocking editorship) and shared authors (interlocking authorship), as well as two forms of intellectual proximity, derived from bibliographic coupling and textual similarity. These four dimensions are integrated using Similarity Network Fusion to produce a unified similarity network from which journal communities are identified. Comparing the field in 2006, 2012, and 2019 reveals a high degree of structural continuity. Our findings suggest that, despite changes in research topics after the crisis, fundamental social and intellectual relationships among journals have remained remarkably stable. Editorial networks, in particular, continue to shape hierarchies and legitimize knowledge production.
LLMs for Productivity
Abstract
Large Language Models (LLMs) have demonstrated remarkable progress in reasoning across diverse domains. However, effective reasoning in real-world tasks requires adapting the reasoning strategy to the demands of the problem, ranging from fast, intuitive responses to deliberate, step-by-step reasoning and tool-augmented thinking. Drawing inspiration from cognitive psychology, we propose a novel taxonomy of LLM reasoning strategies along two knowledge boundaries: a fast/slow boundary separating intuitive from deliberative processes, and an internal/external boundary distinguishing reasoning grounded in the model's parameters from reasoning augmented by external tools. We systematically survey recent work on adaptive reasoning in LLMs and categorize methods based on key decision factors. We conclude by highlighting open challenges and future directions toward more adaptive, efficient, and reliable LLMs.
Large Language Models (LLMs) have demonstrated remarkable progress in reasoning across diverse domains. However, effective reasoning in real-world tasks requires adapting the reasoning strategy to the demands of the problem, ranging from fast, intuitive responses to deliberate, step-by-step reasoning and tool-augmented thinking. Drawing inspiration from cognitive psychology, we propose a novel taxonomy of LLM reasoning strategies along two knowledge boundaries: a fast/slow boundary separating intuitive from deliberative processes, and an internal/external boundary distinguishing reasoning grounded in the model's parameters from reasoning augmented by external tools. We systematically survey recent work on adaptive reasoning in LLMs and categorize methods based on key decision factors. We conclude by highlighting open challenges and future directions toward more adaptive, efficient, and reliable LLMs.
Abstract
Large language models (LLMs) can bias towards relying on their own or the user's information in chat history, leading to overly stubborn or agreeable behaviors in multi-turn conversations. In this paper, we formalize this model characteristic as user-assistant bias and introduce an 8k multi-turn conversation dataset $\textbf{UserAssist}$, which we use to benchmark, understand and manipulate the user-assistant bias in frontier LLMs. Leveraging $\textbf{UserAssist-test}$, we first benchmark the user-assistant bias of 26 commercial and 26 open-weight models. Commercial models show various levels of user bias. Evaluation on open-weight models reveals significant user bias in the instruction-tuned models, and weak user bias in reasoning (or reasoning-distilled) models. We then perform controlled fine-tuning experiments to pinpoint the post-training recipe contributing to these bias shifts: human preference alignment increases user bias, while training on chain-of-thought reasoning traces decreases it. Finally, we demonstrate that user-assistant bias can be bidirectionally adjusted by performing direct preference optimization (DPO) on $\textbf{UserAssist-train}$, and generalizes well to both in-domain and out-of-domain conversations. Our results provide insights into how the LLM integrates information from different sources, and also a viable way to detect and control model abnormalities.
Large language models (LLMs) can bias towards relying on their own or the user's information in chat history, leading to overly stubborn or agreeable behaviors in multi-turn conversations. In this paper, we formalize this model characteristic as user-assistant bias and introduce an 8k multi-turn conversation dataset $\textbf{UserAssist}$, which we use to benchmark, understand and manipulate the user-assistant bias in frontier LLMs. Leveraging $\textbf{UserAssist-test}$, we first benchmark the user-assistant bias of 26 commercial and 26 open-weight models. Commercial models show various levels of user bias. Evaluation on open-weight models reveals significant user bias in the instruction-tuned models, and weak user bias in reasoning (or reasoning-distilled) models. We then perform controlled fine-tuning experiments to pinpoint the post-training recipe contributing to these bias shifts: human preference alignment increases user bias, while training on chain-of-thought reasoning traces decreases it. Finally, we demonstrate that user-assistant bias can be bidirectionally adjusted by performing direct preference optimization (DPO) on $\textbf{UserAssist-train}$, and generalizes well to both in-domain and out-of-domain conversations. Our results provide insights into how the LLM integrates information from different sources, and also a viable way to detect and control model abnormalities.
Help us improve your experience!
This project is on its early stages your feedback can be pivotal on the future of the project.
Let us know what you think about this week's papers and suggestions!
Give Feedback