arXiv subjects

Faisal Farooq

Publications and source records attributed to Faisal Farooq.

5 recordsLinked to original sources

arXiv2026

Measuring the Prevalence of Policy Violating Content with ML Assisted Sampling and LLM Labeling

Content safety teams need metrics that reflect what users actually experience, not only what is reported. We study prevalence: the fraction of user views (impressions) that went to content violating a given policy on a given day. Accurate prevalence measurement is challenging because violations are often rare and human labeling is costly, making frequent, platform-representative studies slow. We present a design-based measurement system that (i) draws daily probability samples from the impression stream using ML-assisted weights to concentrate label budget on high-exposure and high-risk content while preserving unbiasedness, (ii) labels sampled items with a multimodal LLM governed by policy prompts and gold-set validation, and (iii) produces design-consistent prevalence estimates with confidence intervals and dashboard drilldowns. A key design goal is one global sample with many pivots: the same daily sample supports prevalence by surface, viewer geography, content age, and other segments through post-stratified estimation. We describe the statistical estimators, variance and confidence interval construction, label-quality monitoring, and an engineering workflow that makes the system configurable across policies.

cs.LG↗

arXiv2026

Decision Quality Evaluation Framework at Pinterest

Online platforms require robust systems to enforce content safety policies at scale. A critical component of these systems is the ability to evaluate the quality of moderation decisions made by both human agents and Large Language Models (LLMs). However, this evaluation is challenging due to the inherent trade-offs between cost, scale, and trustworthiness, along with the complexity of evolving policies. To address this, we present a comprehensive Decision Quality Evaluation Framework developed and deployed at Pinterest. The framework is centered on a high-trust Golden Set (GDS) curated by subject matter experts (SMEs), which serves as a ground truth benchmark. We introduce an automated intelligent sampling pipeline that uses propensity scores to efficiently expand dataset coverage. We demonstrate the framework's practical application in several key areas: benchmarking the cost-performance trade-offs of various LLM agents, establishing a rigorous methodology for data-driven prompt optimization, managing complex policy evolution, and ensuring the integrity of policy content prevalence metrics via continuous validation. The framework enables a shift from subjective assessments to a data-driven and quantitative practice for managing content safety systems.

stat.AP↗

arXiv2025

Non-Markovian effects on the steady state properties of a damped harmonic oscillator

We analyze the steady-state characteristics of a damped harmonic oscillator (system) in presence of a non-Markovian bath characterized by Lorentzian spectral density. Although Markovian baths presume memoryless dynamics, the introduction of complex temporal connections by a non-Markovian environment radically modifies the dynamics of the system and its steady-state behaviour. We obtain the steady-state Green's functions and correlation functions of the system using the Schwinger-Keldysh formalism. In both rotating and non-rotating wave approximation, we analyzed various emergent properties like effective temperature and distribution function. We also explore the impact of dissipation and non-Markovian bath on the quantum Zeno and anti-Zeno effects. We show that a transition between Zeno to anti-Zeno effect can be tuned by bath spectral width and the strength of dissipation.

quant-ph↗

arXiv2024

Dissipative dynamics of an interacting spin system with collective damping

The competition between Hamiltonian and Lindblad dynamics in quantum systems give rise to non-equillibrium phenomena with no counter part in conventional condensed matter physics. In this paper, we investigate this interplay of dynamics in infinite range Heisenberg model coupled to a non-Markovian bath and subjected to Lindblad dynamics due to spin flipping at a given site. The spin model is bosonized via Holstein-Primakoff transformations and is shown to be valid for narrow range of parameters in the thermodynamic limit. Using Schwinger-Keldysh technique, we derive mean field solution of the model and observe that the system breaks $\mathcal{Z}_2$-symmetry at the transition point. We calculate effective temperature that has linear dependence on the effective system-bath coupling, and is independent of the dissipation rate and cutoff frequency of the bath spectral density. Furthermore, we study the fluctuations over mean field and show that the dissipative spectrum is modified by ${\rm O}(\frac{1}{N})$ correction term which results change in various physically measurable quantities.

quant-ph↗

arXiv2017

A Novel Data-Driven Framework for Risk Characterization and Prediction from Electronic Medical Records: A Case Study of Renal Failure

Electronic medical records (EMR) contain longitudinal information about patients that can be used to analyze outcomes. Typically, studies on EMR data have worked with established variables that have already been acknowledged to be associated with certain outcomes. However, EMR data may also contain hitherto unrecognized factors for risk association and prediction of outcomes for a disease. In this paper, we present a scalable data-driven framework to analyze EMR data corpus in a disease agnostic way that systematically uncovers important factors influencing outcomes in patients, as supported by data and without expert guidance. We validate the importance of such factors by using the framework to predict for the relevant outcomes. Specifically, we analyze EMR data covering approximately 47 million unique patients to characterize renal failure (RF) among type 2 diabetic (T2DM) patients. We propose a specialized L1 regularized Cox Proportional Hazards (CoxPH) survival model to identify the important factors from those available from patient encounter history. To validate the identified factors, we use a specialized generalized linear model (GLM) to predict the probability of renal failure for individual patients within a specified time window. Our experiments indicate that the factors identified via our data-driven method overlap with the patient characteristics recognized by experts. Our approach allows for scalable, repeatable and efficient utilization of data available in EMRs, confirms prior medical knowledge and can generate new hypothesis without expert supervision.

cs.LG↗