arXiv subjects

Muhammad Haseeb

Publications and source records attributed to Muhammad Haseeb.

10 recordsLinked to original sources

arXiv2026

Faithfulness Is Not Free: Auditing Offline KV-Cache Quantization in Retrieval-Augmented Generation

Retrieval-augmented generation systems can precompute and store key-value caches of retrieved documents to avoid re-encoding context at every query. Quantizing these caches further reduces storage, but no prior work asks whether compression damages faithfulness, whether responses remain grounded in the retrieved evidence. Faithfulness and accuracy are not equivalent: a model can produce a correct answer that is no longer supported by the context it was given. We evaluate Qwen2.5-7B-Instruct under INT8 and INT4 quantization on RGB and HotpotQA, measuring both accuracy and faithfulness with a hallucination detector, NLI entailment, and an LLM judge. INT8 is near-lossless across both metrics. INT4 reduces accuracy and, more critically, even among answers that remain factually correct, over 90% of faithfulness changes are negative, i.e., accuracy metrics are blind to this regression. The harm grows under noisy retrieval and with more retrieved chunks. Faithfulness must be audited before compressed caches are deployed.

cs.CL↗

arXiv2026

Probabilistic Fair Ordering of Events

A growing class of applications depends on fair ordering, where events that occur earlier should be processed before later ones. Providing such guarantees is difficult in practice because clock synchronization is inherently imperfect: events generated at different clients within a short time window may carry timestamps that cannot be reliably ordered. Rather than attempting to eliminate synchronization error, we embrace it and establish a probabilistically fair sequencing process. Tommy is a sequencer that uses a statistical model of per-clock synchronization error to compare noisy timestamps probabilistically. Although this enables ordering of two events, the probabilistic comparator is intransitive, making global ordering non-trivial. We address this challenge by mapping the sequencing problem to a classical ranking problem from social choice theory, which offers principled mechanisms for reasoning with intransitive comparisons. Using this formulation, Tommy produces a partial order of events, achieving significantly better fairness than a Spanner TrueTime-based baseline approach.

cs.NI↗

arXiv2025

Mechanistic Evidence for Preserved-but-Misaligned Representations in Non-IID FedAvg

Federated Averaging (FedAvg) often degrades under non-IID client data, but it remains unclear whether this degradation reflects the loss of client-learned representations or a failure to use representations that are still present. We study this question mechanistically in sparse client-trained vision models, using dense-model controls to test whether the observed effects depend on sparsity. Our analysis combines class-specific circuit discovery, linear probing of frozen representations, head-only finetuning, and sparse feature dictionaries. Across CNN and ResNet models on CIFAR-10 and Fashion-MNIST, severe label skew can drive some per-class accuracies near zero even when class-specific internal structure remains recoverable. Linear probes substantially outperform the aggregated classifier, head-only finetuning partially restores accuracy, and USAE transfer reveals a largely shared feature basis between IID and non-IID models. Together, these diagnostics suggest that, in our setting, non-IID FedAvg degradation is not fully explained by representational erasure; it also reflects misalignment between preserved internal structure and the final prediction pathway.

cs.LG↗

arXiv2025

Beyond Lamport, Towards Probabilistic Fair Ordering

A growing class of applications demands \emph{fair ordering} of events, which ensures that events generated earlier are processed before later events. However, achieving such sequencing is challenging due to the inherent errors in clock synchronization: two events at two clients generated close together may have timestamps that cannot be compared confidently. We advocate for an approach that embraces, rather than eliminates, clock synchronization errors. Instead of attempting to remove the error from a timestamp, \systemname{}, our proposed system, leverages a statistical model to compare two noisy timestamps probabilistically by learning per-clock synchronization error distributions. Our preliminary statistical model computes the probability that one event precedes another by only relying on local clocks of clients. This serves as a foundation for a new relation: \emph{likely-happened-before} denoted by $\xrightarrow{p}$ where $p$ represents the probability that an event happened before another. The $\xrightarrow{p}$ relation provides a basis for ordering multiple events which are otherwise considered \emph{concurrent} by Lamport's \emph{happened-before} ($\rightarrow$) relation. We highlight various related challenges including the intransitivity of the $\xrightarrow{p}$ relation as opposed to the transitive $\rightarrow$ relation. We outline several research directions: online fair sequencing, stochastically fair total ordering, and handling byzantine clients.

cs.NI↗

arXiv2025

Context Engineering for Multi-Agent LLM Code Assistants Using Elicit, NotebookLM, ChatGPT, and Claude Code

Large Language Models (LLMs) have shown promise in automating code generation and software engineering tasks, yet they often struggle with complex, multi-file projects due to context limitations and knowledge gaps. We propose a novel context engineering workflow that combines multiple AI components: an Intent Translator (GPT-5) for clarifying user requirements, an Elicit-powered semantic literature retrieval for injecting domain knowledge, NotebookLM-based document synthesis for contextual understanding, and a Claude Code multi-agent system for code generation and validation. Our integrated approach leverages intent clarification, retrieval-augmented generation, and specialized sub-agents orchestrated via Claude's agent framework. We demonstrate that this method significantly improves the accuracy and reliability of code assistants in real-world repositories, yielding higher single-shot success rates and better adherence to project context than baseline single-agent approaches. Qualitative results on a large Next.js codebase show the multi-agent system effectively plans, edits, and tests complex features with minimal human intervention. We compare our system with recent frameworks like CodePlan, MASAI, and HyperAgent, highlighting how targeted context injection and agent role decomposition lead to state-of-the-art performance. Finally, we discuss the implications for deploying LLM-based coding assistants in production, along with lessons learned on context management and future research directions.

cs.SE↗

arXiv2024

Design and Implementation of a Scalable Financial Exchange in the Public Cloud

Financial exchanges are migrating to the cloud, but the best-effort nature of the public cloud is at odds with the stringent latency requirements of exchanges. We present Jasper, a system for meeting the networking requirements of financial exchanges on the public cloud. Jasper uses an overlay tree to scalably multicast market data from an exchange to ~1000 participants with low latency (250 microseconds) and a 1-microsecond difference in data reception time between any two participants. Jasper reuses the same tree for scalable inbound communication (participants to exchange), augmenting it with order pacing and a new priority queue, Limit Order Queue (LOQ), to efficiently handle bursts of market orders. Jasper achieves better scalability and 50% lower latency than the AWS multicast service. During bursty market activity, LOQ nearly doubles the order processing rate.

cs.NI↗

arXiv2022

A Data-Centric Approach to Generate Invariants for a Smart Grid Using Machine Learning

Cyber-Physical Systems (CPS) have gained popularity due to the increased requirements on their uninterrupted connectivity and process automation. Due to their connectivity over the network including intranet and internet, dependence on sensitive data, heterogeneous nature, and large-scale deployment, they are highly vulnerable to cyber-attacks. Cyber-attacks are performed by creating anomalies in the normal operation of the systems with a goal either to disrupt the operation or destroy the system completely. The study proposed here focuses on detecting those anomalies which could be the cause of cyber-attacks. This is achieved by deriving the rules that govern the physical behavior of a process within a plant. These rules are called Invariants. We have proposed a Data-Centric approach (DaC) to generate such invariants. The entire study was conducted using the operational data of a functional smart power grid which is also a living lab.

cs.CR↗

arXiv2021

To Block or Not to Block: Accelerating Mobile Web Pages On-The-Fly Through JavaScript Classification

The increasing complexity of JavaScript in modern mobile web pages has become a critical performance bottleneck for low-end mobile phone users, especially in developing regions. In this paper, we propose SlimWeb, a novel approach that automatically derives lightweight versions of mobile web pages on-the-fly by eliminating the use of unnecessary JavaScript. SlimWeb consists of a JavaScript classification service powered by a supervised Machine Learning (ML) model that provides insights into each JavaScript element embedded in a web page. SlimWeb aims to improve the web browsing experience by predicting the class of each element, such that essential elements are preserved and non-essential elements are blocked by the browsers using the service. We motivate the core design of SlimWeb using a user preference survey of 306 users and perform a detailed evaluation of SlimWeb across 500 popular web pages in a developing region on real 3G and 4G cellular networks, along with a user experience study with 20 real-world users and a usage willingness survey of 588 users. Evaluation results show that SlimWeb achieves a 50% reduction in the page load time compared to the original pages, and more than 30% reduction compared to competing solutions, while achieving high similarity scores to the original pages measured via a qualitative evaluation study of 62 users. SlimWeb improves the overall user experience by more than 60% compared to the original pages, while maintaining 90%-100% of the visual and functional components of most pages. Finally, the SlimWeb classifier achieves a median accuracy of 90% in predicting the JavaScript category.

cs.OH↗

arXiv2021

HiCOPS: High Performance Computing Framework for Tera-Scale Database Search of Mass Spectrometry based Omics Data

Database-search algorithms, that deduce peptides from Mass Spectrometry (MS) data, have tried to improve the computational efficiency to accomplish larger, and more complex systems biology studies. Existing serial, and high-performance computing (HPC) search engines, otherwise highly successful, are known to exhibit poor-scalability with increasing size of theoretical search-space needed for increased complexity of modern non-model, multi-species MS-based omics analysis. Consequently, the bottleneck for computational techniques is the communication costs of moving the data between hierarchy of memory, or processing units, and not the arithmetic operations. This post-Moore change in architecture, and demands of modern systems biology experiments have dampened the overall effectiveness of the existing HPC workflows. We present a novel efficient parallel computational method, and its implementation on memory-distributed architectures for peptide identification tool called HiCOPS, that enables more than 100-fold improvement in speed over most existing HPC proteome database search tools. HiCOPS empowers the supercomputing database search concept for comprehensive identification of peptides, and all their modified forms within a reasonable time-frame. We demonstrate this by searching Gigabytes of experimental MS data against Terabytes of databases where HiCOPS completes peptide identification in few minutes using 72 parallel nodes (1728 cores) compared to several weeks required by existing state-of-the-art tools using 1 node (24 cores); 100 minutes vs 5 weeks; 500x speedup. Finally, we formulate a theoretical framework for our overhead-avoiding strategy, and report superior performance evaluation results for key metrics including execution time, CPU utilization, speedups, and I/O efficiency. The software will be made available at: hicops.github.io

cs.DC↗

arXiv2020

Communication Lower-Bounds for Distributed-Memory Computations for Mass Spectrometry based Omics Data

Mass spectrometry (MS) based omics data analysis require significant time and resources. To date, few parallel algorithms have been proposed for deducing peptides from mass spectrometry-based data. However, these parallel algorithms were designed, and developed when the amount of data that needed to be processed was smaller in scale. In this paper, we prove that the communication bound that is reached by the \emph{existing} parallel algorithms is $\Omega(mn+2r\frac{q}{p})$, where $m$ and $n$ are the dimensions of the theoretical database matrix, $q$ and $r$ are dimensions of spectra, and $p$ is the number of processors. We further prove that communication-optimal strategy with fast-memory $\sqrt{M} = mn + \frac{2qr}{p}$ can achieve $\Omega({\frac{2mnq}{p}})$ but is not achieved by any existing parallel proteomics algorithms till date. To validate our claim, we performed a meta-analysis of published parallel algorithms, and their performance results. We show that sub-optimal speedups with increasing number of processors is a direct consequence of not achieving the communication lower-bounds. We further validate our claim by performing experiments which demonstrate the communication bounds that are proved in this paper. Consequently, we assert that next-generation of \emph{provable}, and demonstrated superior parallel algorithms are urgently needed for MS based large systems-biology studies especially for meta-proteomics, proteogenomic, microbiome, and proteomics for non-model organisms. Our hope is that this paper will excite the parallel computing community to further investigate parallel algorithms for highly influential MS based omics problems.

cs.DC↗