arXiv subjects

Tania Lorido-Botran

Publications and source records attributed to Tania Lorido-Botran.

7 recordsLinked to original sources

arXiv2026

LEVANTE-bench: Multi-Scale Comparison of VLMs to Children Using Cognitive Tasks (or, "Is Your VLM Smarter Than a 5th Grader?")

Given the inherently multimodal nature of human experience, vision-language models (VLMs) hold substantial promise for modeling human cognition as it grows and develops with experience. Realizing their potential requires tools for comparing VLMs with human cognitive development across tasks, ages, and populations. We present LEVANTE-bench, a benchmark based on tasks and data from the Learning Variability Network (LEVANTE), which distributes open-source tasks and data measuring children's cognition across languages and cultures. In LEVANTE-bench, we systematically assess VLMs on six tasks, comparing their alignment with children aged 5-12 ($N$ = 1547) across three countries. We compare models at multiple scales, assessing their overall accuracy, their task- and item-level alignment with children, and how well they match children's trial-level error distributions. Alignment was heterogeneous across scales: at the level of tasks and items, more capable models aligned better with humans. However, match to human error distributions varied widely across tasks, and for several tasks, smaller models matched younger children's errors better. In addition, even the best-performing VLMs struggled on matrix reasoning and mental rotation tasks. Thus, current VLM architectures align only partially with the cognitive abilities of children.

cs.LG↗

arXiv2025

Brook-2PL: Tolerating High Contention Workloads with A Deadlock-Free Two-Phase Locking Protocol

The problem of hotspots remains a critical challenge in high-contention workloads for concurrency control (CC) protocols. Traditional concurrency control approaches encounter significant difficulties under high contention, resulting in excessive transaction aborts and deadlocks. In this paper, we propose Brook-2PL, a novel two-phase locking (2PL) protocol that (1) introduces SLW-Graph for deadlock-free transaction execution, and (2) proposes partial transaction chopping for early lock release. Previous methods suffer from transaction aborts that lead to wasted work and can further burden the system due to their cascading effects. Brook-2PL addresses this limitation by statically analyzing a new graph-based dependency structure called SLW-Graph, enabling deadlock-free two-phase locking through predetermined lock acquisition. Brook-2PL also reduces contention by enabling early lock release using partial transaction chopping and static transaction analysis. We overcome the inherent limitations of traditional transaction chopping by providing a more flexible chopping method. Evaluation using both our synthetic online game store workload and the TPC-C benchmark shows that Brook-2PL significantly outperforms state-of-the-art CC protocols. Brook-2PL achieves an average speed-up of 2.86x while reducing tail latency (p95) by 48% in the TPC-C benchmark.

cs.DB↗

arXiv2025

LLM-Powered Decentralized Generative Agents with Adaptive Hierarchical Knowledge Graph for Cooperative Planning

Developing intelligent agents for long-term cooperation in dynamic open-world scenarios is a major challenge in multi-agent systems. Traditional Multi-agent Reinforcement Learning (MARL) frameworks like centralized training decentralized execution (CTDE) struggle with scalability and flexibility. They require centralized long-term planning, which is difficult without custom reward functions, and face challenges in processing multi-modal data. CTDE approaches also assume fixed cooperation strategies, making them impractical in dynamic environments where agents need to adapt and plan independently. To address decentralized multi-agent cooperation, we propose Decentralized Adaptive Knowledge Graph Memory and Structured Communication System (DAMCS) in a novel Multi-agent Crafter environment. Our generative agents, powered by Large Language Models (LLMs), are more scalable than traditional MARL agents by leveraging external knowledge and language for long-term planning and reasoning. Instead of fully sharing information from all past experiences, DAMCS introduces a multi-modal memory system organized as a hierarchical knowledge graph and a structured communication protocol to optimize agent cooperation. This allows agents to reason from past interactions and share relevant information efficiently. Experiments on novel multi-agent open-world tasks show that DAMCS outperforms both MARL and LLM baselines in task efficiency and collaboration. Compared to single-agent scenarios, the two-agent scenario achieves the same goal with 63% fewer steps, and the six-agent scenario with 74% fewer steps, highlighting the importance of adaptive memory and structured communication in achieving long-term goals. We publicly release our project at: https://happyeureka.github.io/damcs.

cs.AI↗

arXiv2024

IPA: Inference Pipeline Adaptation to Achieve High Accuracy and Cost-Efficiency

Efficiently optimizing multi-model inference pipelines for fast, accurate, and cost-effective inference is a crucial challenge in machine learning production systems, given their tight end-to-end latency requirements. To simplify the exploration of the vast and intricate trade-off space of latency, accuracy, and cost in inference pipelines, providers frequently opt to consider one of them. However, the challenge lies in reconciling latency, accuracy, and cost trade-offs. To address this challenge and propose a solution to efficiently manage model variants in inference pipelines, we present IPA, an online deep learning Inference Pipeline Adaptation system that efficiently leverages model variants for each deep learning task. Model variants are different versions of pre-trained models for the same deep learning task with variations in resource requirements, latency, and accuracy. IPA dynamically configures batch size, replication, and model variants to optimize accuracy, minimize costs, and meet user-defined latency Service Level Agreements (SLAs) using Integer Programming. It supports multi-objective settings for achieving different trade-offs between accuracy and cost objectives while remaining adaptable to varying workloads and dynamic traffic patterns. Navigating a wider variety of configurations allows \namex{} to achieve better trade-offs between cost and accuracy objectives compared to existing methods. Extensive experiments in a Kubernetes implementation with five real-world inference pipelines demonstrate that IPA improves end-to-end accuracy by up to 21% with a minimal cost increase. The code and data for replications are available at https://github.com/reconfigurable-ml-pipeline/ipa.

cs.DC↗

arXiv2023

MSF-Model: Queuing-Based Analysis and Prediction of Metastable Failures in Replicated Storage Systems

Metastable failure is a recent abstraction of a pattern of failures that occurs frequently in real-world distributed storage systems. In this paper, we propose a formal analysis and modeling of metastable failures in replicated storage systems. We focus on a foundational problem in distributed systems -- the problem of consensus -- to have an impact on a large class of systems. Our main contribution is the development of a queuing-based analytical model, MSF-Model, that can be used to characterize and predict metastable failures. MSF-Model integrates novel modeling concepts that allow modeling metastable failures which was interactable to model prior to our work. We also perform real experiments to reproduce and validate our model. Our real experiments show that MSF-Model predicts metastable failures with high accuracy by comparing the real experiment with the predictions from the queuing-based model.

cs.DC↗

arXiv2022

FIRE: A Failure-Adaptive Reinforcement Learning Framework for Edge Computing Migrations

In edge computing, users' service profiles are migrated due to user mobility. Reinforcement learning (RL) frameworks have been proposed to do so, often trained on simulated data. However, existing RL frameworks overlook occasional server failures, which although rare, impact latency-sensitive applications like autonomous driving and real-time obstacle detection. Nevertheless, these failures (rare events), being not adequately represented in historical training data, pose a challenge for data-driven RL algorithms. As it is impractical to adjust failure frequency in real-world applications for training, we introduce FIRE, a framework that adapts to rare events by training a RL policy in an edge computing digital twin environment. We propose ImRE, an importance sampling-based Q-learning algorithm, which samples rare events proportionally to their impact on the value function. FIRE considers delay, migration, failure, and backup placement costs across individual and shared service profiles. We prove ImRE's boundedness and convergence to optimality. Next, we introduce novel deep Q-learning (ImDQL) and actor critic (ImACRE) versions of our algorithm to enhance scalability. We extend our framework to accommodate users with varying risk tolerances. Through trace driven experiments, we show that FIRE reduces costs compared to vanilla RL and the greedy baseline in the event of failures.

cs.NI↗

arXiv2017

PlinyCompute: A Platform for High-Performance, Distributed, Data-Intensive Tool Development

This paper describes PlinyCompute, a system for development of high-performance, data-intensive, distributed computing tools and libraries. In the large, PlinyCompute presents the programmer with a very high-level, declarative interface, relying on automatic, relational-database style optimization to figure out how to stage distributed computations. However, in the small, PlinyCompute presents the capable systems programmer with a persistent object data model and API (the "PC object model") and associated memory management system that has been designed from the ground-up for high performance, distributed, data-intensive computing. This contrasts with most other Big Data systems, which are constructed on top of the Java Virtual Machine (JVM), and hence must at least partially cede performance-critical concerns such as memory management (including layout and de/allocation) and virtual method/function dispatch to the JVM. This hybrid approach---declarative in the large, trusting the programmer's ability to utilize PC object model efficiently in the small---results in a system that is ideal for the development of reusable, data-intensive tools and libraries. Through extensive benchmarking, we show that implementing complex objects manipulation and non-trivial, library-style computations on top of PlinyCompute can result in a speedup of 2x to more than 50x or more compared to equivalent implementations on Spark.

cs.DB↗