Chetan Bansal

Zhaoyang Wang, Qianhui Wu, Xuchao Zhang et al. · microsoft-research

Autonomous web agents powered by large language models (LLMs) have shown promise in completing complex browser tasks, yet they still struggle with long-horizon workflows. A key bottleneck is the grounding gap in existing skill formulations: textual workflow skills provide natural language guidance but cannot be directly executed, while code-based skills are executable but opaque to the agent, offering no step-level understanding for error recovery or adaptation. We introduce WebXSkill, a framework that bridges this gap with executable skills, each pairing a parameterized action program with step-level natural language guidance, enabling both direct execution and agent-driven adaptation. WebXSkill operates in three stages: skill extraction mines reusable action subsequences from readily available synthetic agent trajectories and abstracts them into parameterized skills, skill organization indexes skills into a URL-based graph for context-aware retrieval, and skill deployment exposes two complementary modes, grounded mode for fully automated multi-step execution and guided mode where skills serve as step-by-step instructions that the agent follows with its native planning. On WebArena and WebVoyager, WebXSkill improves task success rate by up to 9.8 and 12.9 points over the baseline, respectively, demonstrating the effectiveness of executable skills for web agents. The code is publicly available at https://github.com/aiming-lab/WebXSkill.

Toufique Ahmed, Supriyo Ghosh, Chetan Bansal et al. · cmu, ibm-research

Incident management for cloud services is a complex process involving several steps and has a huge impact on both service health and developer productivity. On-call engineers require significant amount of domain knowledge and manual effort for root causing and mitigation of production incidents. Recent advances in artificial intelligence has resulted in state-of-the-art large language models like GPT-3.x (both GPT-3.0 and GPT-3.5), which have been used to solve a variety of problems ranging from question answering to text summarization. In this work, we do the first large-scale study to evaluate the effectiveness of these models for helping engineers root cause and mitigate production incidents. We do a rigorous study at Microsoft, on more than 40,000 incidents and compare several large language models in zero-shot, fine-tuned and multi-task setting using semantic and lexical metrics. Lastly, our human evaluation with actual incident owners show the efficacy and future potential of using artificial intelligence for resolving cloud incidents.

Manish Shetty, Chetan Bansal, Sai Pramod Upadhyayula et al. · berkeley

Incident management is a key aspect of operating large-scale cloud services. To aid with faster and efficient resolution of incidents, engineering teams document frequent troubleshooting steps in the form of Troubleshooting Guides (TSGs), to be used by on-call engineers (OCEs). However, TSGs are siloed, unstructured, and often incomplete, requiring developers to manually understand and execute necessary steps. This results in a plethora of issues such as on-call fatigue, reduced productivity, and human errors. In this work, we conduct a large-scale empirical study of over 4K+ TSGs mapped to 1000s of incidents and find that TSGs are widely used and help significantly reduce mitigation efforts. We then analyze feedback on TSGs provided by 400+ OCEs and propose a taxonomy of issues that highlights significant gaps in TSG quality. To alleviate these gaps, we investigate the automation of TSGs and propose AutoTSG -- a novel framework for automation of TSGs to executable workflows by combining machine learning and program synthesis. Our evaluation of AutoTSG on 50 TSGs shows the effectiveness in both identifying TSG statements (accuracy 0.89) and parsing them for execution (precision 0.94 and recall 0.91). Lastly, we survey ten Microsoft engineers and show the importance of TSG automation and the usefulness of AutoTSG.

Zhaoyang Wang, Yiming Liang, Xuchao Zhang et al.

Web agents struggle to adapt to new websites due to the scarcity of environment specific tasks and demonstrations. Recent works have explored synthetic data generation to address this challenge, however, they suffer from data quality issues where synthesized tasks contain hallucinations that cannot be executed, and collected trajectories are noisy with redundant or misaligned actions. In this paper, we propose SynthAgent, a fully synthetic supervision framework that aims at improving synthetic data quality via dual refinement of both tasks and trajectories. Our approach begins by synthesizing diverse tasks through categorized exploration of web elements, ensuring efficient coverage of the target environment. During trajectory collection, we refine tasks when conflicts with actual observations are detected, mitigating hallucinations while maintaining task consistency. After collection, we conduct trajectory refinement with a global context to mitigate potential noise or misalignments. Finally, we fine-tune open-source web agents on the refined synthetic data to adapt them to the target environment. Experimental results demonstrate that SynthAgent outperforms existing synthetic data methods, validating the importance of high-quality synthetic supervision. The code will be publicly available at https://github.com/aiming-lab/SynthAgent.

Manish Shetty, Yinfang Chen, Gagan Somashekar et al.

The rapid growth in the use of Large Language Models (LLMs) and AI Agents as part of software development and deployment is revolutionizing the information technology landscape. While code generation receives significant attention, a higher-impact application lies in using AI agents for operational resilience of cloud services, which currently require significant human effort and domain knowledge. There is a growing interest in AI for IT Operations (AIOps) which aims to automate complex operational tasks, like fault localization and root cause analysis, thereby reducing human intervention and customer impact. However, achieving the vision of autonomous and self-healing clouds through AIOps is hampered by the lack of standardized frameworks for building, evaluating, and improving AIOps agents. This vision paper lays the groundwork for such a framework by first framing the requirements and then discussing design decisions that satisfy them. We also propose AIOpsLab, a prototype implementation leveraging agent-cloud-interface that orchestrates an application, injects real-time faults using chaos engineering, and interfaces with an agent to localize and resolve the faults. We report promising results and lay the groundwork to build a modular and robust framework for building, evaluating, and improving agents for autonomous clouds.

Anish Biswas, Kanishk Goel, Srivarshinee S et al.

Agentic applications are LLMs that iteratively invoke external tools to accomplish complex tasks. Such tool-based agents are rapidly becoming the dominant paradigm for deploying language models in production. Unlike traditional single-turn inference, agentic workloads chain together multiple LLM calls and tool executions before producing a final response, creating a new performance bottleneck that manifests as increased latency in First Token Rendered (FTR) of the final answer. Through analysis of requests at production scale, we reveal three critical challenges: tool calls account for 30-85% of FTR latency, KV cache hit rates collapse despite substantial context reuse across iterations, and sequential orchestration wastes potential intra-request parallelism. These bottlenecks stem from a design gap in which orchestrators and LLM engines operate as decoupled black boxes, preventing cross-layer optimizations. We present Sutradhara, a co-designed agentic inference system that integrates orchestration with LLM serving through a thin API enabling three optimizations: overlap tool execution with subsequent LLM prefill using tool-aware prompt splitting, streaming tool execution to dispatch tools incrementally during decode rather than waiting for complete output, and orchestrator-aware cache management that uses semantic hints to improve hit rates and reduce thrashing. Implemented on vLLM, Sutradhara improves the throughput-latency trade-off in agentic systems, sustains up to 77% higher load at the same median FTR latency, or reduces median FTR latency by up to 15% at the same load while reducing end-to-end latency by upto 11% on A100 GPUs.

Chenyu Zhao, Shenglin Zhang, Zeshun Huang et al.

Large language models (LLMs) have shown growing potential in software engineering, yet few benchmarks evaluate their ability to repair software during migration across instruction set architectures (ISAs). Cross-ISA migration, such as between x86_64 and aarch64, requires handling complex dependencies, heterogeneous toolchains, and long build logs while ensuring executable verification. To address this challenge, we present Build-bench, an end-to-end benchmark that systematically evaluates the capability of LLMs to repair build failures in cross-ISA settings. Build-bench collects 268 real-world failed packages and integrates auxiliary tools including Structure Extraction, File Content Extraction, Content Modification, and Build Verification to support autonomous, tool-augmented reasoning. The repair process operates in an iterative loop where, upon failure, the model receives updated build logs and previous repair outcomes to refine subsequent attempts. Through a comparative evaluation across the studied models, Build-bench reveals that current models achieve a maximum build success rate of 63.19% and tool usage patterns differ significantly across models. By coupling real build environments with verifiable outcomes, Build-bench establishes the first architecture-aware benchmark for studying LLM-based software build and repair.

Dylan Zhang, Xuchao Zhang, Chetan Bansal et al.

Major cloud providers have employed advanced AI-based solutions like large language models to aid humans in identifying the root causes of cloud incidents. Despite the growing prevalence of AI-driven assistants in the root cause analysis process, their effectiveness in assisting on-call engineers is constrained by low accuracy due to the intrinsic difficulty of the task, a propensity for LLM-based approaches to hallucinate, and difficulties in distinguishing these well-disguised hallucinations. To address this challenge, we propose to perform confidence estimation for the predictions to help on-call engineers make decisions on whether to adopt the model prediction. Considering the black-box nature of many LLM-based root cause predictors, fine-tuning or temperature-scaling-based approaches are inapplicable. We therefore design an innovative confidence estimation framework based on prompting retrieval-augmented large language models (LLMs) that demand a minimal amount of information from the root cause predictor. This approach consists of two scoring phases: the LLM-based confidence estimator first evaluates its confidence in making judgments in the face of the current incident that reflects its ``grounded-ness" level in reference data, then rates the root cause prediction based on historical references. An optimization step combines these two scores for a final confidence assignment. We show that our method is able to produce calibrated confidence estimates for predicted root causes, validate the usefulness of retrieved historical data and the prompting strategy as well as the generalizability across different root cause prediction models. Our study takes an important move towards reliably and effectively embedding LLMs into cloud incident management systems.

Zhaoyang Wang, Weilei He, Zhiyuan Liang et al.

Recent self-rewarding large language models (LLM) have successfully applied LLM-as-a-Judge to iteratively improve the alignment performance without the need of human annotations for preference data. These methods commonly utilize the same LLM to act as both the policy model (which generates responses) and the reward model (which scores and ranks those responses). The ranked responses are then used as preference pairs to train the LLM via direct alignment technologies (e.g. DPO). However, it is noteworthy that throughout this process, there is no guarantee of accuracy in the rewarding and ranking, which is critical for ensuring accurate rewards and high-quality preference data. Empirical results from relatively small LLMs (e.g., 7B parameters) also indicate that improvements from self-rewarding may diminish after several iterations in certain situations, which we hypothesize is due to accumulated bias in the reward system. This bias can lead to unreliable preference data for training the LLM. To address this issue, we first formulate and analyze the generalized iterative preference fine-tuning framework for self-rewarding language model. We then introduce the regularization to this generalized framework to mitigate the overconfident preference labeling in the self-rewarding process. Based on this theoretical insight, we propose a Consistency Regularized sElf-rewarding lAnguage Model (CREAM) that leverages the consistency of rewards across different iterations to regularize the self-rewarding training, helping the model to learn from more reliable preference data. With this explicit regularization, our empirical results demonstrate the superiority of CREAM in improving both reward consistency and alignment performance. The code is publicly available at https://github.com/Raibows/CREAM.

Menglin Xia, Xuchao Zhang, Shantanu Dixit et al.

Agent memory systems must accommodate continuously growing information while supporting efficient, context-aware retrieval for downstream tasks. Abstraction is essential for scaling agent memory, yet it often comes at the cost of specificity, obscuring the fine-grained details required for effective reasoning. We introduce Memora, a harmonic memory representation that structurally balances abstraction and specificity. Memora organizes information via its primary abstractions that index concrete memory values and consolidate related updates into unified memory entries, while cue anchors expand retrieval access across diverse aspects of the memory and connect related memories. Building on this structure, we employ a retrieval policy that actively exploits these memory connections to retrieve relevant information beyond direct semantic similarity. Theoretically, we show that standard Retrieval-Augmented Generation (RAG) and Knowledge Graph (KG)-based memory systems emerge as special cases of our framework. Empirically, Memora establishes a new state-of-the-art on the LoCoMo and LongMemEval benchmarks, demonstrating better retrieval relevance and reasoning effectiveness as memory scales.

Zhenfeng Lin, Haoji Hu, Ming Hao et al.

Outage management in large-scale cloud operations remains heavily manual, requiring rapid triage, cross-team coordination, and experience-driven decisions under partial observability. We present \textbf{ActionNex}, a production-grade agentic system that supports end-to-end outage assistance, including real-time updates, knowledge distillation, and role- and stage-conditioned next-best action recommendations. ActionNex ingests multimodal operational signals (e.g., outage content, telemetry, and human communications) and compresses them into critical events that represent meaningful state transitions. It couples this perception layer with a hierarchical memory subsystem: long-term Key-Condition-Action (KCA) knowledge distilled from playbooks and historical executions, episodic memory of prior outages, and working memory of the live context. A reasoning agent aligns current critical events to preconditions, retrieves relevant memories, and generates actionable recommendations; executed human actions serve as an implicit feedback signal to enable continual self-evolution in a human-agent hybrid system. We evaluate ActionNex on eight real Azure outages (8M tokens, 4,000 critical events) using two complementary ground-truth action sets, achieving 71.4\% precision and 52.8-54.8\% recall. The system has been piloted in production and has received positive early feedback.

Shraddha Barke, Arnav Goyal, Alind Khare et al.

AI agents often fail in ways that are difficult to localize because executions are probabilistic, long-horizon, multi-agent, and mediated by noisy tool outputs. We address this gap by manually annotating failed agent runs and release a novel benchmark of 115 failed trajectories spanning structured API workflows, incident management, and open-ended web/file tasks. Each trajectory is annotated with a critical failure step and a category from a grounded-theory derived, cross-domain failure taxonomy. To mitigate the human cost of failure attribution, we present AGENTRX, an automated domain-agnostic diagnostic framework that pinpoints the critical failure step in a failed agent trajectory. It synthesizes constraints, evaluates them step-by-step, and produces an auditable validation log of constraint violations with associated evidence; an LLM-based judge uses this log to localize the critical step and category. Our framework improves step localization and failure attribution over existing baselines across three domains.

Tianrun Yu, Yuxiao Yang, Zhaoyang Wang et al.

We study test-time scaling, where a model improves its answer through multi-round self-reflection at inference. We introduce In-Context Policy Optimization (ICPO), in which an agent optimizes its response in context using self-assessed or externally observed rewards without modifying its parameters. To explain this ICPO process, we theoretically show that with sufficient pretraining under a novel Fisher-weighted logit-matching objective, a single-layer linear self-attention model can provably imitate policy-optimization algorithm for linear bandits. Building on this theory, we propose Minimum-Entropy ICPO (ME-ICPO), a practical algorithm that iteratively uses its response and self-assessed reward to refine its response in-context at inference time. By selecting the responses and their rewards with minimum entropy, ME-ICPO ensures the robustness of the self-assessed rewards via majority voting. Across standard mathematical reasoning tasks, ME-ICPO attains competitive, top-tier performance while keeping inference costs affordable compared with other inference-time algorithms. Overall, ICPO provides a principled understanding of self-reflection in LLMs and yields practical benefits for test-time scaling for mathematical reasoning.

Zhaoyang Wang, Jinqi Jiang, Huichi Zhou et al.

Recent Large Language Models (LLMs) have demonstrated satisfying general instruction following ability. However, small LLMs with about 7B parameters still struggle fine-grained format following (e.g., JSON format), which seriously hinder the advancements of their applications. Most existing methods focus on benchmarking general instruction following while overlook how to improve the specific format following ability for small LLMs. Besides, these methods often rely on evaluations based on advanced LLMs (e.g., GPT-4), which can introduce the intrinsic bias of LLMs and be costly due to the API calls. In this paper, we first curate a fully verifiable format following dataset VFF. In contrast to existing works often adopting external LLMs for instruction-following validations, every sample of VFF can be easily validated with a Python function. Further, we propose to leverage this verifiable feature to synthesize massive data for progressively training small LLMs, in order to improve their format following abilities. Experimental results highlight the prevalent limitations in the format following capabilities of 7B level open-source LLMs and demonstrate the effectiveness of our method in enhancing this essential ability.

Haoran Qiu, Anish Biswas, Zihan Zhao et al.

Large multimodal models (LMMs) demonstrate impressive capabilities in understanding images, videos, and audio beyond text. However, efficiently serving LMMs in production environments poses significant challenges due to their complex architectures and heterogeneous characteristics across their multi-stage inference pipelines. We present the first comprehensive systems analysis of two prominent LMM architectures, decoder-only and cross-attention, across six representative open-source models, revealing key systems design implications. We also present an in-depth analysis of production LMM inference traces, uncovering unique workload characteristics, including variable, heavy-tailed request distributions and bursty traffic patterns. Based on these insights, we propose ModServe, a modular LMM serving system that decouples stages for independent optimization and adaptive scaling. ModServe dynamically reconfigures stages and handles bursty traffic with modality-aware scheduling and autoscaling to meet tail latency SLOs while minimizing costs. ModServe achieves 3.3-5.5x higher throughput (leading to 25-41.3% cost saving) while meeting SLOs on a 128-GPU cluster with production traces.

Sidharth Sinha, Anson Bastos, Xuchao Zhang et al.

Large language models (LLMs) excel in open domains but struggle in specialized settings with limited data and evolving knowledge. Existing domain adaptation practices rely heavily on manual trial-and-error processes, incur significant hyperparameter complexity, and are highly sensitive to data and user preferences, all under the high cost of LLM training. Moreover, the interactions and transferability of hyperparameter choices across models/domains remain poorly understood, making adaptation gains uncertain even with substantial effort. To solve these challenges, we present AutoAdapt, a novel end-to-end automated framework for efficient and reliable LLM domain adaptation. AutoAdapt leverages curated knowledge bases from literature and open-source resources to reduce expert intervention. To narrow the search space, we design a novel multi-agent debating system in which proposal and critic agents iteratively interact to align user intent and incorporate data signals and best practices into the planning process. To optimize hyperparameters under tight budgets, we propose AutoRefine, a novel LLM-based surrogate that replaces costly black-box search. Across 10 tasks, AutoAdapt achieves a 25% average relative accuracy improvement over state-of-the-art Automated Machine Learning baselines with minimal overhead.

Taneesh Gupta, Rahul Madhavan, Xuchao Zhang et al.

Multi-preference optimization enriches language-model alignment beyond pairwise preferences by contrasting entire sets of helpful and undesired responses, thereby enabling richer training signals for large language models. During self-play alignment, these models often produce numerous candidate answers per query, rendering it computationally infeasible to include all responses in the training objective. In this work, we propose $\textit{Active Multi-Preference Optimization}$ (AMPO), a novel approach that combines on-policy generation, a multi-preference group-contrastive loss, and active subset selection. Specifically, we score and embed large candidate pools of responses and then select a small, yet informative, subset that covers reward extremes and distinct semantic clusters for preference optimization. Our contrastive training scheme is capable of identifying not only the best and worst answers but also subtle, underexplored modes that are crucial for robust alignment. Theoretically, we provide guarantees for expected reward maximization using our active selection method, and empirically, AMPO achieves state-of-the-art results on $\textit{AlpacaEval}$ using Llama 8B and Mistral 7B. We release our datasets $\href{https://huggingface.co/Multi-preference-Optimization}{here}$.

Chenyu Zhao, Minghua Ma, Shenglin Zhang et al.

Frequent toolchain updates and growing ISA diversity have made system-level software package repair increasingly important. Diagnosing and repairing build failures remains challenging because failures involve heterogeneous evidence, dependency constraints, and architecture-specific build conventions. While recent LLM-based repair methods show promise for project-level source fixes, they struggle with system-level repair, where failures span multi-language artifacts such as build recipes, scripts, and source archives, and require iterative validation through external build services. In this paper, we first conduct a systematic empirical study of real-world system-level build failures. We find that 72% of failures stem from dependency and environment misconfigurations rather than isolated code defects, suggesting that effective repair must prioritize packaging logic and iterative feedback. Motivated by these insights, we propose EvidenT, an evidence-preserving repair framework that decouples iteration-aware evidence management from tool execution. EvidenT includes: (1) an external Build Service for reproducible execution and feedback; (2) an Evidence-Preserving Repair Controller that fuses repair history, knowledge context, and build artifacts; and (3) an automated Repair Orchestrator that invokes modular tools for failure localization and system-level repair in a closed-loop validation environment. We evaluate EvidenT on 219 real-world RISC-V package build failures. EvidenT repairs 118 packages (53.88%), outperforming state-of-the-art agentic baselines (20.55%) and direct LLM-based repair (1.83%). To assess architectural generality, we extend EvidenT to legacy ISAs by updating only ISA-specific knowledge context. Preliminary experiments achieve success rates of 41.77% on aarch64 and 46.99% on x86_64, demonstrating robustness across diverse hardware ecosystems.

Chenyu Zhao, Shenglin Zhang, Yihang Lin et al.

Software engineering agents are increasingly deployed in evaluable engineering environments, yet post-failure recovery remains costly, manual, and ad hoc. Existing systems expose traces or generate follow-up feedback, but they do not convert heterogeneous runtime evidence into grounded, bounded recovery guidance for a subsequent attempt. We present PROBE, a failure-anchored framework for structured recovery in software engineering agents. PROBE organizes failed-run telemetry into structured evidence, structured diagnosis, and bounded recovery guidance through a Telemetry Layer, a Diagnosis Layer, and a Guidance Gate. The Telemetry Layer preserves fine-grained runtime signals, the Diagnosis Layer fuses cross-signal evidence into grounded diagnoses, and the Guidance Gate produces diagnosis-derived guidance only when it is evidence-grounded, actionable, and within the scope of agent-side behavior. We evaluate PROBE across three settings: repository-level software repair, enterprise workflow recovery, and AIOps service mitigation. On 257 initially unresolved cases, PROBE achieves 65.37% Top-1 diagnosis accuracy and a 21.79% recovery rate, outperforming the strongest non-PROBE baseline by 43.58 and 12.45 percentage points. The results reveal a diagnosis-recovery gap: accurate diagnosis is necessary but insufficient unless translated into bounded guidance that a subsequent attempt can execute and verify. Beyond controlled evaluation, a Microsoft IcM prototype shows that PROBE can attach as a non-intrusive side channel to existing service-diagnosis workflows without changing the agent policy, toolset, or execution budget. These results suggest that telemetry-grounded, failure-anchored recovery can improve post-failure recoverability under realistic engineering constraints.

Devjeet Roy, Xuchao Zhang, Rashi Bhave et al.

The growing complexity of cloud based software systems has resulted in incident management becoming an integral part of the software development lifecycle. Root cause analysis (RCA), a critical part of the incident management process, is a demanding task for on-call engineers, requiring deep domain knowledge and extensive experience with a team's specific services. Automation of RCA can result in significant savings of time, and ease the burden of incident management on on-call engineers. Recently, researchers have utilized Large Language Models (LLMs) to perform RCA, and have demonstrated promising results. However, these approaches are not able to dynamically collect additional diagnostic information such as incident related logs, metrics or databases, severely restricting their ability to diagnose root causes. In this work, we explore the use of LLM based agents for RCA to address this limitation. We present a thorough empirical evaluation of a ReAct agent equipped with retrieval tools, on an out-of-distribution dataset of production incidents collected at Microsoft. Results show that ReAct performs competitively with strong retrieval and reasoning baselines, but with highly increased factual accuracy. We then extend this evaluation by incorporating discussions associated with incident reports as additional inputs for the models, which surprisingly does not yield significant performance improvements. Lastly, we conduct a case study with a team at Microsoft to equip the ReAct agent with tools that give it access to external diagnostic services that are used by the team for manual RCA. Our results show how agents can overcome the limitations of prior work, and practical considerations for implementing such a system in practice.

Jun Liu, Chaoyun Zhang, Jiaxu Qian et al.

Time series anomaly detection (TSAD) plays a crucial role in various industries by identifying atypical patterns that deviate from standard trends, thereby maintaining system integrity and enabling prompt response measures. Traditional TSAD models, which often rely on deep learning, require extensive training data and operate as black boxes, lacking interpretability for detected anomalies. To address these challenges, we propose LLMAD, a novel TSAD method that employs Large Language Models (LLMs) to deliver accurate and interpretable TSAD results. LLMAD innovatively applies LLMs for in-context anomaly detection by retrieving both positive and negative similar time series segments, significantly enhancing LLMs' effectiveness. Furthermore, LLMAD employs the Anomaly Detection Chain-of-Thought (AnoCoT) approach to mimic expert logic for its decision-making process. This method further enhances its performance and enables LLMAD to provide explanations for their detections through versatile perspectives, which are particularly important for user decision-making. Experiments on three datasets indicate that our LLMAD achieves detection performance comparable to state-of-the-art deep learning methods while offering remarkable interpretability for detections. To the best of our knowledge, this is the first work that directly employs LLMs for TSAD.

Drishti Goel, Fiza Husain, Aditya Singh et al.

Incident management for large cloud services is a complex and tedious process and requires significant amount of manual efforts from on-call engineers (OCEs). OCEs typically leverage data from different stages of the software development lifecycle [SDLC] (e.g., codes, configuration, monitor data, service properties, service dependencies, trouble-shooting documents, etc.) to generate insights for detection, root causing and mitigating of incidents. Recent advancements in large language models [LLMs] (e.g., ChatGPT, GPT-4, Gemini) created opportunities to automatically generate contextual recommendations to the OCEs assisting them to quickly identify and mitigate critical issues. However, existing research typically takes a silo-ed view for solving a certain task in incident management by leveraging data from a single stage of SDLC. In this paper, we demonstrate that augmenting additional contextual data from different stages of SDLC improves the performance of two critically important and practically challenging tasks: (1) automatically generating root cause recommendations for dependency failure related incidents, and (2) identifying ontology of service monitors used for automatically detecting incidents. By leveraging 353 incident and 260 monitor dataset from Microsoft, we demonstrate that augmenting contextual information from different stages of the SDLC improves the performance over State-of-The-Art methods.

Yinfang Chen, Manish Shetty, Gagan Somashekar et al.

AI for IT Operations (AIOps) aims to automate complex operational tasks, such as fault localization and root cause analysis, to reduce human workload and minimize customer impact. While traditional DevOps tools and AIOps algorithms often focus on addressing isolated operational tasks, recent advances in Large Language Models (LLMs) and AI agents are revolutionizing AIOps by enabling end-to-end and multitask automation. This paper envisions a future where AI agents autonomously manage operational tasks throughout the entire incident lifecycle, leading to self-healing cloud systems, a paradigm we term AgentOps. Realizing this vision requires a comprehensive framework to guide the design, development, and evaluation of these agents. To this end, we present AIOPSLAB, a framework that not only deploys microservice cloud environments, injects faults, generates workloads, and exports telemetry data but also orchestrates these components and provides interfaces for interacting with and evaluating agents. We discuss the key requirements for such a holistic framework and demonstrate how AIOPSLAB can facilitate the evaluation of next-generation AIOps agents. Through evaluations of state-of-the-art LLM agents within the benchmark created by AIOPSLAB, we provide insights into their capabilities and limitations in handling complex operational tasks in cloud environments.

Yiyang Zhou, Zhaoyang Wang, Tianle Wang et al.

High-quality preference data is essential for aligning foundation models with human values through preference learning. However, manual annotation of such data is often time-consuming and costly. Recent methods often adopt a self-rewarding approach, where the target model generates and annotates its own preference data, but this can lead to inaccuracies since the reward model shares weights with the target model, thereby amplifying inherent biases. To address these issues, we propose Anyprefer, a framework designed to synthesize high-quality preference data for aligning the target model. Anyprefer frames the data synthesis process as a cooperative two-player Markov Game, where the target model and the judge model collaborate together. Here, a series of external tools are introduced to assist the judge model in accurately rewarding the target model's responses, mitigating biases in the rewarding process. In addition, a feedback mechanism is introduced to optimize prompts for both models, enhancing collaboration and improving data quality. The synthesized data is compiled into a new preference dataset, Anyprefer-V1, consisting of 58K high-quality preference pairs. Extensive experiments show that Anyprefer significantly improves model alignment performance across four main applications, covering 21 datasets, achieving average improvements of 18.55% in five natural language generation datasets, 3.66% in nine vision-language understanding datasets, 30.05% in three medical image analysis datasets, and 16.00% in four visuo-motor control tasks.

Pooja Srinivas, Fiza Husain, Anjaly Parayil et al.

Cloud service owners need to continuously monitor their services to ensure high availability and reliability. Gaps in monitoring can lead to delay in incident detection and significant negative customer impact. Current process of monitor creation is ad-hoc and reactive in nature. Developers create monitors using their tribal knowledge and, primarily, a trial and error based process. As a result, monitors often have incomplete coverage which leads to production issues, or, redundancy which results in noise and wasted effort. In this work, we address this issue by proposing an intelligent monitoring framework that recommends monitors for cloud services based on their service properties. We start by mining the attributes of 30,000+ monitors from 791 production services at Microsoft and derive a structured ontology for monitors. We focus on two crucial dimensions: what to monitor (resources) and which metrics to monitor. We conduct an extensive empirical study and derive key insights on the major classes of monitors employed by cloud services at Microsoft, their associated dimensions, and the interrelationship between service properties and this ontology. Using these insights, we propose a deep learning based framework that recommends monitors based on the service properties. Finally, we conduct a user study with engineers from Microsoft which demonstrates the usefulness of the proposed framework. The proposed framework along with the ontology driven projections, succeeded in creating production quality recommendations for majority of resource classes. This was also validated by the users from the study who rated the framework's usefulness as 4.27 out of 5.

Taneesh Gupta, Rahul Madhavan, Xuchao Zhang et al.

Direct Preference Optimization (DPO) has become a popular approach for aligning language models using pairwise preferences. However, in practical post-training pipelines, on-policy generation typically yields multiple candidate responses per prompt, which are scored by a reward model to guide learning. In this setting, we propose $\textbf{Multi-Preference Optimization (MPO)}$, a generalization of DPO that optimizes over entire sets of responses by extending the Bradley-Terry model to groupwise comparisons between chosen and rejected sets. To further enhance learning, MPO employs deviation-based weighting, which emphasizes outlier responses that deviate most from the mean reward, effectively inducing a self-paced curriculum. We theoretically prove that MPO reduces alignment bias at a rate of $\mathcal{O}\left(\frac{1}{\sqrt{n}}\right)$ with respect to the number of responses per query. Empirically, MPO achieves state-of-the-art performance on the UltraFeedback benchmark and yields up to $\sim 17.5\%$ improvement over the state-of-the-art baseline in length-controlled win rate on AlpacaEval2, establishing a new baseline for preference-based alignment

Sukrit Kumar, Drishti Goel, Thomas Zimmermann et al.

Software developers balance a variety of different tasks in a workweek, yet the allocation of time often differs from what they consider ideal. Identifying and addressing these deviations is crucial for organizations aiming to enhance the productivity and well-being of the developers. In this paper, we present the findings from a survey of 484 software developers at Microsoft, which aims to identify the key differences between how developers would like to allocate their time during an ideal workweek versus their actual workweek. Our analysis reveals significant deviations between a developer's ideal workweek and their actual workweek, with a clear correlation: as the gap between these two workweeks widens, we observe a decline in both productivity and satisfaction. By examining these deviations in specific activities, we assess their direct impact on the developers' satisfaction and productivity. Additionally, given the growing adoption of AI tools in software engineering, both in the industry and academia, we identify specific tasks and areas that could be strong candidates for automation. In this paper, we make three key contributions: 1) We quantify the impact of workweek deviations on developer productivity and satisfaction 2) We identify individual tasks that disproportionately affect satisfaction and productivity 3) We provide actual data-driven insights to guide future AI automation efforts in software engineering, aligning them with the developers' requirements and ideal workflows for maximizing their productivity and satisfaction.

Supriyo Ghosh, Karish Grover, Jimmy Wong et al.

Despite significant reliability efforts, large-scale cloud services inevitably experience production incidents that can significantly impact service availability and customer's satisfaction. Worse, in many cases one incident can lead to multiple downstream failures due to cascading effects that creates several related incidents across different dependent services. Often time On-call Engineers (OCEs) examine these incidents in silos that lead to significant amount of manual toil and increase the overall time-to-mitigate incidents. Therefore, developing efficient incident linking models is of paramount importance for grouping related incidents into clusters so as to quickly resolve major outages and reduce on-call fatigue. Existing incident linking methods mostly leverages textual and contextual information of incidents (e.g., title, description, severity, impacted components), thus failing to leverage the inter-dependencies between services. In this paper, we propose the dependency-aware incident linking (DiLink) framework which leverages both textual and service dependency graph information to improve the accuracy and coverage of incident links not only coming from same service, but also from different services and workloads. Furthermore, we propose a novel method to align the embeddings of multi-modal (i.e., textual and graphical) data using Orthogonal Procrustes. Extensive experimental results on real-world incidents from 5 workloads of Microsoft demonstrate that our alignment method has an F1-score of 0.96 (14% gain over current state-of-the-art methods). We are also in the process of deploying this solution across 610 services from these 5 workloads for continuously supporting OCEs improving incident management and reducing manual toil.

Redwan Ibne Seraj Khan, Kunal Jain, Haiying Shen et al.

In a multi-tenant large language model (LLM) serving platform hosting diverse applications, some users may submit an excessive number of requests, causing the service to become unavailable to other users and creating unfairness. Existing fairness approaches do not account for variations in token lengths across applications and multiple LLM calls, making them unsuitable for such platforms. To address the fairness challenge, this paper analyzes millions of requests from thousands of users on MS CoPilot, a real-world multi-tenant LLM platform hosted by Microsoft. Our analysis confirms the inadequacy of existing methods and guides the development of FairServe, a system that ensures fair LLM access across diverse applications. FairServe proposes application-characteristic aware request throttling coupled with a weighted service counter based scheduling technique to curb abusive behavior and ensure fairness. Our experimental results on real-world traces demonstrate FairServe's superior performance compared to the state-of-the-art method in ensuring fairness. We are actively working on deploying our system in production, expecting to benefit millions of customers world-wide.

Taneesh Gupta, Shivam Shandilya, Xuchao Zhang et al.

Reward modeling in large language models is susceptible to reward hacking, causing models to latch onto superficial features such as the tendency to generate lists or unnecessarily long responses. In reinforcement learning from human feedback (RLHF) and more generally during post-training flawed reward signals often lead to outputs that optimize for these spurious correlates instead of genuine quality or correctness. We propose Context-Aware Reward Modeling (CARMO), a novel approach that first generates dynamic, context-relevant criteria to ground the reward model before producing reward scores. Unlike prior methods that rely on static rubrics, CARMO leverages large language models (LLMs) to adaptively create evaluation criteria such as logical consistency, clarity, and depth tailored to the user query. Our theoretical analysis shows that such criteria generation can mitigate reward hacking. We further demonstrate that CARMO can be distilled into smaller models, reducing the computational cost of alignment. We establish a new state-of-the-art performance in zero-shot settings for generative models, achieving a 2.1\% improvement on Reward Bench. Furthermore, alignment performed on the CARMO-curated preference dataset achieves 22.5\% and 21.1\% LC-WR and WR, respectively, on Mistral-Base (7B).

Lu Wang, Mayukh Das, Fangkai Yang et al.

We address the challenge of learning safe and robust decision policies in presence of uncertainty in context of the real scientific problem of adaptive resource oversubscription to enhance resource efficiency while ensuring safety against resource congestion risk. Traditional supervised prediction or forecasting models are ineffective in learning adaptive policies whereas standard online optimization or reinforcement learning is difficult to deploy on real systems. Offline methods such as imitation learning (IL) are ideal since we can directly leverage historical resource usage telemetry. But, the underlying aleatoric uncertainty in such telemetry is a critical bottleneck. We solve this with our proposed novel chance-constrained imitation learning framework, which ensures implicit safety against uncertainty in a principled manner via a combination of stochastic (chance) constraints on resource congestion risk and ensemble value functions. This leads to substantial ($\approx 3-4\times$) improvement in resource efficiency and safety in many oversubscription scenarios, including resource management in cloud services.

Anson Bastos, Shreeya Venneti, Anjaly Parayil et al.

Reliability of large-scale cloud services is critical for user satisfaction and business continuity. Despite significant investments in reliability engineering, production incidents remain inevitable, often leading to customer impact and operational overhead. In large cloud companies, multiple services are deployed across regions necessitating robust health monitoring systems. However, the current monitor configuration process is manual, largely reactive and ad hoc, resulting in gaps in coverage and redundant alerts. In this paper, we present a comprehensive study of monitor creation in Microsoft, identifying key components in the existing process. We further design a modular recommendation framework that processes the graph structured service entities to suggest optimal monitor configurations. Through extensive experimentation on historical data and user study of recommendations for production services at Microsoft, we demonstrate the efficacy of our approach in providing relevant recommendations for monitor configurations.

Shangzhe Li, Xuchao Zhang, Chetan Bansal et al.

Self-play post-training methods has emerged as an effective approach for finetuning large language models and turn the weak language model into strong language model without preference data. However, the theoretical foundations for self-play finetuning remain underexplored. In this work, we tackle this by connecting self-play finetuning with adversarial imitation learning by formulating finetuning procedure as a min-max game between the model and a regularized implicit reward player parameterized by the model itself. This perspective unifies self-play imitation and general preference alignment within a common framework. Under this formulation, we present a game-theoretic analysis showing that the self-play finetuning will converge to it's equilibrium. Guided by this theoretical formulation, we propose a new self-play imitation finetuning algorithm based on the $χ^2$-divergence variational objective with bounded rewards and improved stability. Experiments on various of language model finetuning tasks demonstrate consistent improvements over existing self-play methods and validate our theoretical insights.

Shashwat Jaiswal, Shrikara Arun, Anjaly Parayil et al.

Low-Rank Adaptation (LoRA) has become the de facto method for parameter-efficient fine-tuning of large language models (LLMs), enabling rapid adaptation to diverse domains. In production, LoRA-based models are served at scale, creating multi-tenant environments with hundreds of adapters sharing a base model. However, state-of-the-art serving systems co-batch heterogeneous adapters without accounting for rank (size) variability, leading to severe performance skew, which ultimately requires adding more GPUs to satisfy service-level objectives (SLOs). Existing optimizations, focused on loading, caching, and kernel execution, ignore this heterogeneity, leaving GPU resources underutilized. We present LoRAServe, a workload-aware dynamic adapter placement and routing framework designed to tame rank diversity in LoRA serving. By dynamically rebalancing adapters across GPUs and leveraging GPU Direct RDMA for remote access, LoRAServe maximizes throughput and minimizes tail latency under real-world workload drift. Evaluations on production traces from Company X show that LoRAServe elicits up to 2$\times$ higher throughput, up to 9$\times$ lower TTFT, while using up to 50% fewer GPUs under SLO constraints compared to state-of-the-art systems.

Fiza Hussain, Anson Bastos, Anjaly Parayil et al.

In this paper, we present DiRecGNN, an attention-enhanced entity recommendation framework for monitoring cloud services at Microsoft. We provide insights on the usefulness of this feature as perceived by the cloud service owners and lessons learned from deployment. Specifically, we introduce the problem of recommending the optimal subset of attributes (dimensions) that should be tracked by an automated watchdog (monitor) for cloud services. To begin, we construct the monitor heterogeneous graph at production-scale. The interaction dynamics of these entities are often characterized by limited structural and engagement information, resulting in inferior performance of state-of-the-art approaches. Moreover, traditional methods fail to capture the dependencies between entities spanning a long range due to their homophilic nature. Therefore, we propose an attention-enhanced entity ranking model inspired by transformer architectures. Our model utilizes a multi-head attention mechanism to focus on heterogeneous neighbors and their attributes, and further attends to paths sampled using random walks to capture long-range dependencies. We also employ multi-faceted loss functions to optimize for relevant recommendations while respecting the inherent sparsity of the data. Empirical evaluations demonstrate significant improvements over existing methods, with our model achieving a 43.1% increase in MRR. Furthermore, product teams who consumed these features perceive the feature as useful and rated it 4.5 out of 5.

Yizhu Jiao, Xuchao Zhang, Zhaoyang Wang et al.

Current Large Language Models (LLMs) excel in general reasoning yet struggle with specialized tasks requiring proprietary or domain-specific knowledge. Fine-tuning large models for every niche application is often infeasible due to black-box constraints and high computational overhead. To address this, we propose a collaborative framework that pairs a specialized weak model with a general strong model. The weak model, tailored to specific domains, produces initial drafts and background information, while the strong model leverages its advanced reasoning to refine these drafts, extending LLMs' capabilities to critical yet specialized tasks. To optimize this collaboration, we introduce a collaborative feedback to fine-tunes the weak model, which quantifies the influence of the weak model's contributions in the collaboration procedure and establishes preference pairs to guide preference tuning of the weak model. We validate our framework through experiments on three domains. We find that the collaboration significantly outperforms each model alone by leveraging complementary strengths. Moreover, aligning the weak model with the collaborative preference further enhances overall performance.

Taneesh Gupta, Rahul Madhavan, Xuchao Zhang et al.

To mitigate reward hacking from response verbosity, modern preference optimization methods are increasingly adopting length normalization (e.g., SimPO, ORPO, LN-DPO). While effective against this bias, we demonstrate that length normalization itself introduces a failure mode: the URSLA shortcut. Here models learn to satisfy the alignment objective by prematurely truncating low-quality responses rather than learning from their semantic content. To address this, we introduce REFA, a new alignment framework that proposes probabilistic control on a structural token that controls termination. Our core innovation is a new class of regularizers that operate directly on the probability of the End-of-Sequence (EOS) token, a previously unexploited control lever. This token-level intervention provides a principled solution to the URSLA shortcut, ensuring genuine quality improvements. Furthermore, it unlocks a versatile mechanism for managing the alignment-efficiency tradeoff, enabling practitioners to fine-tune models that adhere to specific token budgets. Empirically, REFA achieves a 60.29% win rate and a 52.17% length-controlled win rate on AlpacaEval2 with Llama-3-8B-Instruct, demonstrating the power of our token-level control paradigm.

Aditya Soni, Mayukh Das, Anjaly Parayil et al.

The difficulty of exploring and training online on real production systems limits the scope of real-time online data/feedback-driven decision making. The most feasible approach is to adopt offline reinforcement learning from limited trajectory samples. However, after deployment, such policies fail due to exogenous factors that temporarily or permanently disturb/alter the transition distribution of the assumed decision process structure induced by offline samples. This results in critical policy failures and generalization errors in sensitive domains like Real-Time Communication (RTC). We solve this crucial problem of identifying robust actions in presence of domain shifts due to unseen exogenous stochastic factors in the wild. As it is impossible to learn generalized offline policies within the support of offline data that are robust to these unseen exogenous disturbances, we propose a novel post-deployment shaping of policies (Streetwise), conditioned on real-time characterization of out-of-distribution sub-spaces. This leads to robust actions in bandwidth estimation (BWE) of network bottlenecks in RTC and in standard benchmarks. Our extensive experimental results on BWE and other standard offline RL benchmark environments demonstrate a significant improvement ($\approx$ 18% on some scenarios) in final returns wrt. end-user metrics over state-of-the-art baselines.

Peng Xia, Ze Chen, Juanxi Tian et al.

Artificial intelligence has significantly impacted medical applications, particularly with the advent of Medical Large Vision Language Models (Med-LVLMs), sparking optimism for the future of automated and personalized healthcare. However, the trustworthiness of Med-LVLMs remains unverified, posing significant risks for future model deployment. In this paper, we introduce CARES and aim to comprehensively evaluate the Trustworthiness of Med-LVLMs across the medical domain. We assess the trustworthiness of Med-LVLMs across five dimensions, including trustfulness, fairness, safety, privacy, and robustness. CARES comprises about 41K question-answer pairs in both closed and open-ended formats, covering 16 medical image modalities and 27 anatomical regions. Our analysis reveals that the models consistently exhibit concerns regarding trustworthiness, often displaying factual inaccuracies and failing to maintain fairness across different demographic groups. Furthermore, they are vulnerable to attacks and demonstrate a lack of privacy awareness. We publicly release our benchmark and code in https://cares-ai.github.io/.

Xuchao Zhang, Supriyo Ghosh, Chetan Bansal et al.

Root Cause Analysis (RCA) plays a pivotal role in the incident diagnosis process for cloud services, requiring on-call engineers to identify the primary issues and implement corrective actions to prevent future recurrences. Improving the incident RCA process is vital for minimizing service downtime, customer impact and manual toil. Recent advances in artificial intelligence have introduced state-of-the-art Large Language Models (LLMs) like GPT-4, which have proven effective in tackling various AIOps problems, ranging from code authoring to incident management. Nonetheless, the GPT-4 model's immense size presents challenges when trying to fine-tune it on user data because of the significant GPU resource demand and the necessity for continuous model fine-tuning with the emergence of new data. To address the high cost of fine-tuning LLM, we propose an in-context learning approach for automated root causing, which eliminates the need for fine-tuning. We conduct extensive study over 100,000 production incidents, comparing several large language models using multiple metrics. The results reveal that our in-context learning approach outperforms the previous fine-tuned large language models such as GPT-3 by an average of 24.8\% across all metrics, with an impressive 49.7\% improvement over the zero-shot model. Moreover, human evaluation involving actual incident owners demonstrates its superiority over the fine-tuned model, achieving a 43.5\% improvement in correctness and an 8.7\% enhancement in readability. The impressive results demonstrate the viability of utilizing a vanilla GPT model for the RCA task, thereby avoiding the high computational and maintenance costs associated with a fine-tuned model.

Chandra Maddila, Suhas Shanbhogue, Apoorva Agrawal et al.

Software development is information-dense knowledge work that requires collaboration with other developers and awareness of artifacts such as work items, pull requests, and files. With the speed of development increasing, information overload is a challenge for people developing and maintaining these systems. Finding information and people is difficult for software engineers, especially when they work in large software systems or have just recently joined a project. In this paper, we build a large scale data platform named Nalanda platform, which contains two subsystems: 1. A large scale socio-technical graph system, named Nalanda graph system 2. A large scale recommendation system, named Nalanda index system that aims at satisfying the information needs of software developers. The Nalanda graph is an enterprise scale graph with data from 6,500 repositories, with 37,410,706 nodes and 128,745,590 edges. On top of the Nalanda graph system, we built software analytics applications including a newsfeed named MyNalanda, and based on organic growth alone, it has Daily Active Users (DAU) of 290 and Monthly Active Users (MAU) of 590. A preliminary user study shows that 74% of developers and engineering managers surveyed are favorable toward continued use of the platform for information discovery. The Nalanda index system constitutes two indices: artifact index and expert index. It uses the socio-technical graph (Nalanda graph system) to rank the results and provide better recommendations to software developers. A large scale quantitative evaluation shows that the Nalanda index system provides recommendations with an accuracy of 78% for the top three recommendations.

Manish Shetty, Chetan Bansal, Suman Nath et al.

Crash localization, an important step in debugging crashes, is challenging when dealing with an extremely large number of diverse applications and platforms and underlying root causes. Large-scale error reporting systems, e.g., Windows Error Reporting (WER), commonly rely on manually developed rules and heuristics to localize blamed frames causing the crashes. As new applications and features are routinely introduced and existing applications are run under new environments, developing new rules and maintaining existing ones become extremely challenging. We propose a data-driven solution to address the problem. We start with the first large-scale empirical study of 362K crashes and their blamed methods reported to WER by tens of thousands of applications running in the field. The analysis provides valuable insights on where and how the crashes happen and what methods to blame for the crashes. These insights enable us to develop DeepAnalyze, a novel multi-task sequence labeling approach for identifying blamed frames in stack traces. We evaluate our model with over a million real-world crashes from four popular Microsoft applications and show that DeepAnalyze, trained with crashes from one set of applications, not only accurately localizes crashes of the same applications, but also bootstraps crash localization for other applications with zero to very little additional training data.

Manish Shetty, Chetan Bansal, Sumit Kumar et al.

The move from boxed products to services and the widespread adoption of cloud computing has had a huge impact on the software development life cycle and DevOps processes. Particularly, incident management has become critical for developing and operating large-scale services. Prior work on incident management has heavily focused on the challenges with incident triaging and de-duplication. In this work, we address the fundamental problem of structured knowledge extraction from service incidents. We have built SoftNER, a framework for mining Knowledge Graphs from incident reports. First, we build a novel multi-task learning based BiLSTM-CRF model which leverages not just the semantic context but also the data-types for extracting factual information in the form of named entities. Next, we present an approach to mine relations between the named entities for automatically constructing knowledge graphs. We have deployed SoftNER at Microsoft, a major cloud service provider and have evaluated it on more than 2 months of cloud incidents. We show that the unsupervised machine learning pipeline has a high precision of 0.96. Our multi-task learning based deep learning model also outperforms the state-of-the-art NER models. Lastly, using the knowledge extracted by SoftNER, we are able to build accurate models for applications such as incident triaging and recommending entities based on their relevance to incident titles.

Chandra Maddila, Sai Surya Upadrasta, Chetan Bansal et al.

Pull requests are a key part of the collaborative software development and code review process today. However, pull requests can also slow down the software development process when the reviewer(s) or the author do not actively engage with the pull request. In this work, we design an end-to-end service, Nudge, for accelerating overdue pull requests towards completion by reminding the author or the reviewer(s) to engage with their overdue pull requests. First, we use models based on effort estimation and machine learning to predict the completion time for a given pull request. Second, we use activity detection to filter out pull requests that may be overdue, but for which sufficient action is taking place nonetheless. Lastly, we use actor identification to understand who the blocker of the pull request is and nudge the appropriate actor (author or reviewer(s)). The key novelty of Nudge is that it succeeds in reducing pull request resolution time, while ensuring that developers perceive the notifications sent as useful, at the scale of thousands of repositories. In a randomized trial on 147 repositories in use at Microsoft, Nudge was able to reduce pull request resolution time by 60% for 8,500 pull requests, when compared to overdue pull requests for which Nudge did not send a notification. Furthermore, developers receiving Nudge notifications resolved 73% of these notifications as positive. We observed similar results when scaling up the deployment of Nudge to 8,000 repositories at Microsoft, for which Nudge sent 210,000 notifications during a full year. This demonstrates Nudge's ability to scale to thousands of repositories. Lastly, our qualitative analysis of a selection of Nudge notifications indicates areas for future research, such as taking dependencies among pull requests and developer availability into account.

Nikitha Rao, Chetan Bansal, Joe Guan

Developers use search for various tasks such as finding code, documentation, debugging information, etc. In particular, web search is heavily used by developers for finding code examples and snippets during the coding process. Recently, natural language based code search has been an active area of research. However, the lack of real-world large-scale datasets is a significant bottleneck. In this work, we propose a weak supervision based approach for detecting code search intent in search queries for C# and Java programming languages. We evaluate the approach against several baselines on a real-world dataset comprised of over 1 million queries mined from Bing web search engine and show that the CNN based model can achieve an accuracy of 77% and 76% for C# and Java respectively. Furthermore, we are also releasing Search4Code, the first large-scale real-world dataset of code search queries mined from Bing web search engine. We hope that the dataset will aid future research on code search.

Manish Shetty, Chetan Bansal, Sumit Kumar et al.

In the last decade, two paradigm shifts have reshaped the software industry - the move from boxed products to services and the widespread adoption of cloud computing. This has had a huge impact on the software development life cycle and the DevOps processes. Particularly, incident management has become critical for developing and operating large-scale services. Incidents are created to ensure timely communication of service issues and, also, their resolution. Prior work on incident management has been heavily focused on the challenges with incident triaging and de-duplication. In this work, we address the fundamental problem of structured knowledge extraction from service incidents. We have built SoftNER, a framework for unsupervised knowledge extraction from service incidents. We frame the knowledge extraction problem as a Named-entity Recognition task for extracting factual information. SoftNER leverages structural patterns like key,value pairs and tables for bootstrapping the training data. Further, we build a novel multi-task learning based BiLSTM-CRF model which leverages not just the semantic context but also the data-types for named-entity extraction. We have deployed SoftNER at Microsoft, a major cloud service provider and have evaluated it on more than 2 months of cloud incidents. We show that the unsupervised machine learning based approach has a high precision of 0.96. Our multi-task learning based deep learning model also outperforms the state of the art NER models. Lastly, using the knowledge extracted by SoftNER we are able to build significantly more accurate models for important downstream tasks like incident triaging.

Foyzul Hassan, Chetan Bansal, Nachiappan Nagappan et al.

Software engineers spend a substantial amount of time using Web search to accomplish software engineering tasks. Such search tasks include finding code snippets, API documentation, seeking help with debugging, etc. While debugging a bug or crash, one of the common practices of software engineers is to search for information about the associated error or exception traces on the internet. In this paper, we analyze query logs from a leading commercial general-purpose search engine (GPSE) such as Google, Yahoo! or Bing to carry out a large scale study of software exceptions. To the best of our knowledge, this is the first large scale study to analyze how Web search is used to find information about exceptions. We analyzed about 1 million exception related search queries from a random sample of 5 billion web search queries. To extract exceptions from unstructured query text, we built a novel and high-performance machine learning model with a F1-score of 0.82. Using the machine learning model, we extracted exceptions from raw queries and performed popularity, effort, success, query characteristic and web domain analysis. We also performed programming language-specific analysis to give a better view of the exception search behavior. These techniques can help improve existing methods, documentation and tools for exception analysis and prediction. Further, similar techniques can be applied for APIs, frameworks, etc.

Nikitha Rao, Chetan Bansal, Subhabrata Mukherjee et al.

Web search engines are frequently used to access information about products. This has increased in recent times with the rising popularity of e-commerce. However, there is limited understanding of what users search for and their intents when it comes to product search on the web. In this work, we study search logs from Bing web search engine to characterize user intents and study user behavior for product search. We propose a taxonomy of product intents by analyzing product search queries. This is a challenging task given that only 15%-17% of web search queries are about products. We train machine learning classifiers with query log features to classify queries based on intent with an overall F1-score of 78%. We further analyze various characteristics of product search queries in terms of search metrics like dwell time, success, popularity and session-specific information.

Chetan Bansal, Pantazis Deligiannis, Chandra Maddila et al.

Cyber attacks are increasingly becoming prevalent and causing significant damage to individuals, businesses and even countries. In particular, ransomware attacks have grown significantly over the last decade. We do the first study on mining insights about ransomware attacks by analyzing query logs from Bing web search engine. We first extract ransomware related queries and then build a machine learning model to identify queries where users are seeking support for ransomware attacks. We show that user search behavior and characteristics are correlated with ransomware attacks. We also analyse trends in the temporal and geographical space and validate our findings against publicly available information. Lastly, we do a case study on 'Nemty', a popular ransomware, to show that it is possible to derive accurate insights about cyber attacks by query log analysis.

Nikitha Rao, Chetan Bansal, Thomas Zimmermann et al.

Web search plays an integral role in software engineering (SE) to help with various tasks such as finding documentation, debugging, installation, etc. In this work, we present the first large-scale analysis of web search behavior for SE tasks using the search query logs from Bing, a commercial web search engine. First, we use distant supervision techniques to build a machine learning classifier to extract the SE search queries with an F1 score of 93%. We then perform an analysis on one million search sessions to understand how software engineering related queries and sessions differ from other queries and sessions. Subsequently, we propose a taxonomy of intents to identify the various contexts in which web search is used in software engineering. Lastly, we analyze millions of SE queries to understand the distribution, search metrics and trends across these SE search intents. Our analysis shows that SE related queries form a significant portion of the overall web search traffic. Additionally, we found that there are six major intent categories for which web search is used in software engineering. The techniques and insights can not only help improve existing tools but can also inspire the development of new tools that aid in finding information for SE related tasks.