Related papers: Feature Toggle Dynamics in Large-Scale Systems: Pr…
Large language models, LLMs, are increasingly deployed in multiturn settings where earlier responses shape later ones, making reliability dependent on whether a conversation remains consistent over time. When this consistency degrades…
To improve the quality of experience (QoE), especially under moderate to high traffic demand, it is important to understand the impact of the network and application QoS on user experience. This paper comparatively evaluates the impact of…
(1) Background: Container orchestration frameworks provide support for management of complex distributed applications. Different frameworks have emerged only recently, and they have been in constant evolution as new features are being…
Persistence has proved to be a valuable tool to analyze real world data robustly. Several approaches to persistence have been attempted over time, some topological in flavor, based on the vector space-valued homology functor, other…
In search and recommendation systems, predictive models often suffer from temporal instability when certain input features introduce volatility in output scores. This instability can degrade model reliability and user experience especially…
Large language models (LLMs) can carry out human-like dialogue, but unlike humans, they are stateless due to the superposition property. However, during multi-turn, multi-agent interactions, LLMs begin to exhibit consistent, character-like…
Tabular data is widely utilized in various machine learning tasks. Current tabular learning research predominantly focuses on closed environments, while in real-world applications, open environments are often encountered, where distribution…
In Large Language Model (LLM) serving, the KV-cache (KVC) bottleneck causes high tail Time-to-First-Token (TTFT) and Time-Between-Tokens (TBT), impairing user experience, particularly in time-sensitive applications. However, satisfying both…
Background: Machine Learning (ML) systems rely on data to make predictions, the systems have many added components compared to traditional software systems such as the data processing pipeline, serving pipeline, and model training. Existing…
Feature interaction is a core ingredient in ranking models for large-scale recommender systems, yet making it both expressive and efficiently scalable remains challenging. Exhaustive pairwise interaction is powerful but incurs quadratic…
Technical debt refers to the trade-offs between code quality and faster delivery, impacting future development with increased complexity, bugs, and costs. This study empirically analyzes the additional work effort caused by technical debt…
The explorative and iterative nature of developing and operating machine learning (ML) applications leads to a variety of artifacts, such as datasets, features, models, hyperparameters, metrics, software, configurations, and logs. In order…
Companies increasingly use either manual or automated system testing to ensure the quality of their software products. As a system evolves and is extended with new features the test suite also typically grows as new test cases are added. To…
Context: GitHub hosts an impressive number of high-quality OSS projects. However, selecting "the right tool for the job" is a challenging task, because we do not have precise information about those high-quality projects. Objective: In this…
As the capability frontier of autonomous agents continues to expand, they are increasingly able to complete specialized tasks through plug-and-play external skills. Yet current benchmarks mostly test whether models can use provided skills,…
Models of parallel processing systems typically assume that one has $l$ workers and jobs are split into an equal number of $k=l$ tasks. Splitting jobs into $k > l$ smaller tasks, i.e. using ``tiny tasks'', can yield performance and…
Transition Matching (TM) is an emerging paradigm for generative modeling that generalizes diffusion and flow-matching models as well as continuous-state autoregressive models. TM, similar to previous paradigms, gradually transforms noise…
When deploying LLMs in agentic architectures requiring real-time decisions under temporal constraints, we assume they reliably determine whether action windows remain open or have closed. This assumption is untested. We characterize…
Modern software development demands code that is maintainable, testable, and scalable by organizing the implementation into modular components with iterative reuse of existing codes. We formalize this iterative, multi-turn paradigm as…
Deep video recognition is more computationally expensive than image recognition, especially on large-scale datasets like Kinetics [1]. Therefore, training scalability is essential to handle a large amount of videos. In this paper, we study…