Large language models (LLMs) frequently prioritize parametric world knowledge over provided context -- a failure mode that is particularly catastrophic in enterprise or counterfactual settings where local facts contradict web-scale training data. While modern reasoning models improve general response quality, they fail to resolve these underlying prior knowledge biases even when generating a high volume of costly thinking tokens. We propose a context-editing framework that addresses this by performing selective abstraction over entities that appear in both the context and the question. Our approach replaces these overlapping entities with typed, indexed identifiers (e.g., Paris with City łangle1\rangle), suppressing spurious lexical associations while preserving the relational structure required for grounded reasoning. This mechanism is decoupled into an offline preprocessing stage and a lightweight inference-time substitution, requiring no model retraining. Experiments across counterfactual and multi-hop knowledge-conflict benchmarks demonstrate consistent accuracy gains across many model families, open- and closed-sourced, especially for small to medium sized LLMs (0.5B to 18B). Notably, our framework yields up to a 25% improvement over state-of-the-art instruct models and 13% over reasoning models, establishing symbolic abstraction as a highly cost-efficient solution for ensuring context fidelity in LLMs.

We present an unsupervised method for aggregating anomalies in tabular datasets by identifying the top-k tabular data quality insights. Each insight consists of a set of anomalous attributes and the corresponding subsets of records that serve as evidence to the user. The process of identifying these insight blocks is challenging due to (i) the absence of labeled anomalies, (ii) the exponential size of the subset search space, and (iii) the complex dependencies among attributes, which obscure the true sources of anomalies. Simple frequency-based methods fail to capture these dependencies, leading to inaccurate results. To address this, we introduce Tab-Shapley, a cooperative game theory based framework that uses Shapley values to quantify the contribution of each attribute to the data's anomalous nature. While calculating Shapley values typically requires exponential time, we show that our game admits a closed-form solution, making the computation efficient. We validate the effectiveness of our approach through empirical analysis on real-world tabular datasets with ground-truth anomaly labels.

Finding the root cause of failures is a prominent problem in many complex networks. Causal inference provides us with tools to address this problem algorithmically to automate this process and solve it efficiently. The existing methods either use a known causal structure to identify root cause by backtracking the changes, or ignore the causal structure but relies on invariance tests to identify the changing causal mechanisms after the failure. Assuming a single, unknown root cause, we first establish a novel connection between root cause analysis and the \textit{Interactive Graph Search (IGS)} problem. This mapping highlights the importance of causal knowledge: we demonstrate that any algorithm relying solely on marginal invariance tests to identify the root cause must perform at least $\Omega(\log_{2}(n) + d\log_{1+d}n)$ many tests, where $n$ represents the number of components and $d$ denotes the maximum out-degree of the graph. We then present an optimal algorithm that achieves this bound by reducing the root cause identification problem as an instance of IGS. Beyond the single root cause scenario, we propose a practical extension for settings with multiple root causes and partial causal knowledge. More specifically, we show that even if the causal graph is partially known, we can identify the root-causes with a linear number of invariance tests. This is the first known result on incorporating a partial causal structure for root cause analysis. Our experiments on a production-level application demonstrate that, even in the absence of complete causal information, our approach accurately identifies the root causes of failures.

Retrieval-Augmented Generation (RAG) is often used with Large Language Models (LLMs) to infuse domain knowledge or user-specific information. In RAG, given a user query, a retriever extracts chunks of relevant text from a knowledge base. These chunks are sent to an LLM as part of the input prompt. Typically, any given chunk is repeatedly retrieved across user questions. However, currently, for every question, attention layers in LLMs fully compute the Keys and Values (KVs) repeatedly for the input chunks, as state-of-the-art methods cannot reuse KV-caches when chunks appear at arbitrary locations or with arbitrary contexts. Naive reuse leads to output quality degradation. This leads to potentially redundant computations on expensive GPUs and increases latency. In this work, we propose Cache-Craft , a system for managing and reusing precomputed KVs corresponding to the text chunks (which we call chunk-caches ) in RAG-based systems. We present how to identify chunk-caches that are reusable, how to efficiently perform a small fraction of recomputation to fix the cache and maintain output quality, and how to efficiently store and evict chunk-caches in the hardware for maximizing reuse while masking any overheads. With real production workloads as well as synthetic datasets, we show that Cache-Craft reduces redundant computation by 51% over SOTA prefix-caching and 75% over full recomputation. Additionally, with continuous batching on a real production workload, we get a 1.6× speed up in throughput for both the LLama-3-8B and 70B models and a 2.1× and 2× reduction in end-to-end response latency respectively, compared to prefix-caching, while maintaining generation quality.

We study the problem of optimizing data storage and access costs on the cloud while ensuring that the desired performance or latency is unaffected. We first propose an optimizer that optimizes the data placement tier (on the cloud) and the choice of compression schemes to apply, for given data partitions with temporal access predictions. Secondly, we propose a model to learn the compression performance of multiple algorithms across data partitions in different formats to generate compression performance predictions on the fly, as inputs to the optimizer. Thirdly, we propose to approach the data partitioning problem fundamentally differently than the current default in most data lakes where partitioning is in the form of ingestion batches. We propose access pattern aware data partitioning and formulate an optimization problem that optimizes the size and reading costs of partitions subject to access patterns.We study the various optimization problems theoretically as well as empirically, and provide theoretical bounds as well as hardness results. We propose a unified pipeline of cost minimization, called SCOPe that combines the different modules. We extensively compare the performance of our methods with related baselines from the literature on TPC-H data as well as enterprise datasets (ranging from GB to PB in volume) and show that SCOPe substantially improves over the baselines. We show significant cost savings compared to platform baselines, of the order of 50% to 83% on enterprise Data Lake datasets that range from terabytes to petabytes in volume.

Modern cloud services are prone to failures due to their complex architecture, making diagnosis a critical process. Site Reliability Engineers (SREs) spend hours leveraging multiple sources of data, including the alerts, error logs, and domain expertise through past experiences to locate the root cause(s). These experiences are documented as natural language text in outage reports for previous outages. However, utilizing the raw yet rich semi-structured information in the reports systematically is time-consuming. Structured information, on the other hand, such as alerts that are often used during fault diagnosis, is voluminous and requires expert knowledge to discern. Several strategies have been proposed to use each source of data separately for root cause analysis. In this work, we build a diagnostic service called ESRO that recommends root causes and remediation for failures by utilizing structured as well as semi-structured sources of data systematically. ESRO constructs a causal graph using alerts and a knowledge graph using outage reports, and merges them in a novel way to form a unified graph during training. A retrieval based mechanism is then used to search the unified graph and rank the likely root causes and remediation techniques based on the alerts fired during an outage at inference time. Not only the individual alerts, but their respective importance in predicting an outage group is taken into account during recommendation. We evaluated our model on several cloud service outages of a large SaaS enterprise over the course of ~2 years, and obtained an average improvement of 27% in rouge scores after comparing the likely root causes against the ground truth over state-of-the-art baselines. We further establish the effectiveness of ESRO through qualitative analysis on multiple real outage examples.

Cloud services are omnipresent and critical cloud service failure is a fact of life. In order to retain customers and prevent revenue loss, it is important to provide high reliability guarantees for these services. One way to do this is by predicting outages in advance, which can help in reducing the severity as well as time to recovery. It is difficult to forecast critical failures due to the rarity of these events. Moreover, critical failures are ill-defined in terms of observable data. Our proposed method, Outage-Watch, defines critical service outages as deteriorations in the Quality of Service (QoS) captured by a set of metrics. Outage-Watch detects such outages in advance by using current system state to predict whether the QoS metrics will cross a threshold and initiate an extreme event. A mixture of Gaussian is used to model the distribution of the QoS metrics for flexibility and an extreme event regularizer helps in improving learning in tail of the distribution. An outage is predicted if the probability of any one of the QoS metrics crossing threshold changes significantly. Our evaluation on a real-world SaaS company dataset shows that Outage-Watch significantly outperforms traditional methods with an average AUC of 0.98. Additionally, Outage-Watch detects all the outages exhibiting a change in service metrics and reduces the Mean Time To Detection (MTTD) of outages by up to 88% when deployed in an enterprise cloud-service system, demonstrating efficacy of our proposed method.

AI-based monitoring has become crucial for cloud-based services due to its scale. A common approach to AI-based monitoring is to detect causal relationships among service components and build a causal graph. Availability of domain information makes cloud systems even better suited for such causal detection approaches. In modern cloud systems, however, auto-scalers dynamically change the number of microservice instances, and a load-balancer manages the load on each instance. This poses a challenge for off-the-shelf causal structure detection techniques as they neither incorporate the system architectural domain information nor provide a way to model distributed compute across varying numbers of service instances. To address this, we develop CausIL, which detects a causal structure among service metrics by considering compute distributed across dynamic instances and incorporating domain knowledge derived from system architecture. Towards the application in cloud systems, CausIL estimates a causal graph using instance-specific variations in performance metrics, modeling multiple instances of a service as independent, conditional on system assumptions. Simulation study shows the efficacy of CausIL over baselines by improving graph estimation accuracy by ∼ 25% as measured by Structural Hamming Distance whereas the real-world dataset demonstrates CausIL’s applicability in deployment settings.

Most cloud applications use a large number of smaller sub-components (called microservices) that interact with each other in the form of a complex graph to provide the overall functionality to the user. While the modularity of the microservice architecture is beneficial for rapid software development, maintaining and debugging such a system quickly in cases of failure is challenging. We propose a scalable algorithm for rapidly detecting the root cause of failures in complex microservice architectures. The key ideas behind our novel hierarchical and localized learning approach are: (1) to treat the failure as an intervention on the root cause to quickly detect it, (2) only learn the portion of the causal graph related to the root cause, thus avoiding a large number of costly conditional independence tests, and (3) hierarchically explore the graph. The proposed technique is highly scalable and produces useful insights about the root cause, while the use of traditional techniques becomes infeasible due to high computation time. Our solution is application agnostic and relies only on the data collected for diagnosis. For the evaluation, we compare the proposed solution with a modified version of the PC algorithm and the state-of-the-art for root cause analysis. The results show a considerable improvement in top-$k$ recall while significantly reducing the execution time.

Efficient outage detection and remediation is crucial for effectively operating cloud computing systems. To remediate outages, system engineers must quickly identify the causal relationships between metrics and correlate events across multiple monitoring tools. In practice, this process largely remains reactive due to the complexity and general lack of interpretability within such monitoring environments. This work presents ViSRE: an integrated visual analytics system that integrates causal and predictive models with interactive visualizations to aid in proactive cloud outage management. We develop enhanced node representations for our causal graph representation to support system engineers in performing root cause analysis and reasoning about causality chains in multi-dimensional temporal data. We report the results of a quantitative assessment of the proposed predictive models, which show good performance guarantees. To evaluate and refine our system, we conduct a study with six cloud system engineers who verify that our proposed techniques can support proactive cloud maintenance by intuitively displaying temporal relationships between predicted and raw data. By correlating and presenting data from disparate sources, ViSRE also reduces context switching costs and reduces the time spent on manually correlating events during remediation of time-critical outages.

Aggregate advertising-presenting a single ad to large groups of individuals through traditional media such as television and print-presents a unique challenge to measuring efficacy because treatment and outcome are observed from two disparate sources (interaction and revenue realization). In this work, we propose a Bayesian model to estimate the impact of an ad on observable web metrics that are readily available in many modern analytics suites. The proposed model controls for three sources of possible confounding: the time, geography, and content of the advertisement. The proposed model is easily applicable to a wide variety of problems and readily generates error bounds for the estimates. We evaluate our approach on a real dataset for a set of TV ads for an advertiser.

Causal inference using observational data on multiple treatments is an important problem in a wide variety of fields. However, the existing literature tends to focus only on causal inference in case of binary or multinoulli treatments. These models are either incompatible with multiple treatments, or extending them to multiple treatments is computationally expensive. We use a previous formulation of causal inference using variational autoencoder (VAE) and propose a novel architecture to estimate the causal effect of any subset of the treatments. The higher order effects of multiple treatments are captured through a task embedding. The task embedding allows the model to scale to multiple treatments. The model is applied on real digital marketing dataset to evaluate the next best set of marketing actions. For evaluation, the model is compared against competitive baseline models on two semi-synthetic datasets created using the covariates from the real dataset. The performance is measured along four evaluation metrics considered in the causal inference literature and one proposed by us. The proposed evaluation metric measures the loss in the expected outcome when a particular model is used for decision making as compared to the ground truth. The proposed model outperforms the baselines along all five evaluation metrics. It outperforms the best baseline by over 30% along these evaluation metrics. The proposed approach is also shown to be robust when a subset of the confounders is not observed. The results on real data show the importance of the flexible modeling approach provided by the proposed model.

Anomaly detection and forecasting are two fundamental problems in time series analysis that are relevant to a wide range of academic and industrial disciplines. Although these problems have been investigated in the literature previously, the assumptions therein are too restrictive for autonomous analysis. Common examples of limiting assumptions include perfect knowledge about the time series seasonality and/or presence of anomaly (spikes and level changes) free time windows. Current practice is to manually input this knowledge into anomaly detection and forecasting systems which negate any possibility of autonomous analysis. This paper relaxes these assumptions by jointly estimating the latent components (viz. seasonality, level changes, and spikes) in the observed time series without assuming the availability of anomaly-free time windows. The novel and flexible two stage approach proposed herein is based on (a) sparse modeling of the different latent components of the time series and (b) ARMA modeling for fitting the error. The approach leads to a solution for anomaly detection with control over type-I errors. Further, by design, the method is robust against anomalies in the observation window when it is used to solve the forecasting problem by extrapolation. Experiments are conducted with both synthetic and real datasets to demonstrate the efficacy of the proposed method. We compare our approach to various popular baselines. The presented approach outperforms baseline algorithms for anomaly detection in all our experiments and performs favorably for the forecasting task.