arXiv subjects

Harsh Patel

Publications and source records attributed to Harsh Patel.

At least 19 recordsLinked to original sources

arXiv2026

A Verifiable Search Is Not a Learnable Chain-of-Thought

It is tempting to assume any task solvable by a short program can be taught to a model as its chain-of-thought: write the steps out, fine-tune, and the model follows. This paper shows the assumption fails for an identifiable class of procedures. The testbed is nine reasoning tasks, each from a deterministic generator; public and hidden splits share generators, so held-out data proxies test accuracy. I reverse-engineer the generators into Python solvers, render them as chain-of-thought, and distill into a rank-<= 32 LoRA over a 30B (3.5B-active) Nemotron model. Forward-computable tasks install readily: lookup/arithmetic and an 8-bit boolean task transfer (>= 0.99 and 0.68). Cryptarithm does not: distilling its backtracking search holds at 0.01-0.07 across eleven chain-of-thought designs, RL from verifiable rewards, and self-training, even though a search solver answers 71% of instances. This is not a capability gap. The model does the arithmetic on 97-100% of lines and ranks the correct cipher in its top eight on 71%; it cannot carry the search forward as a left-to-right derivation. Fine-tuning learns the shape of a verifiable elimination step while its verdicts become unconditional templates, correct only 16-57% of the time ("verdict-as-token"). The ceiling holds across backbones from 3B to 671B and across fine-tuning and prompting; a controlled intervention isolates the cause: revealing the cipher key, which turns the derivation forward, lifts the same instances from 0.03 to 0.57. When a procedure's only solution is search over information-free structure, no faithful forward chain-of-thought exists to imitate. The task becomes learnable only by removing the search, precomputing its combinatorial core into a catalog and reducing the trace to recall plus verification; the 1st-place solution reaches Private LB 0.92 this way. What distills is memorization and verification, not search.

cs.LG↗

arXiv2026

CEI: A Clonal Expansion Identifier for T-cell receptor clones following SARS-CoV-2 vaccination

Each T cell typically carries a specific T-cell receptor (TCR) that determines its specificity against an epitope presented by the HLA complex on a target cell. Antigenic challenge triggers the expansion of reactive cells within a diverse pool of T cells with randomly generated receptors, a process that results in epitope-driven shifts of TCR frequencies over time. Here, we analyze the effects of SARS-CoV-2 vaccination on the TCR populations in peripheral blood drawn from seven COVID-naive individuals, before vaccines were widely available. To identify SARS-CoV-2 vaccine-associated TCR sequences among the $\sim 10^{5}-10^{6}$ TCR sequences sampled before and after vaccination, we develop statistical criteria to detect significant increases in abundance of positive TCR clones. Application of our statistical methods shows a robust identification of TCR sequences that respond to SARS-CoV-2 vaccination in vivo, illustrating the feasibility of quantifying the clone-specific dynamics of T-cell abundance changes following immunological perturbations.

q-bio.QM↗

arXiv2025

AnyTask: an Automated Task and Data Generation Framework for Advancing Sim-to-Real Policy Learning

Generalist robot learning remains constrained by data: large-scale, diverse, and high-quality interaction data are expensive to collect in the real world. While simulation has become a promising way for scaling up data collection, the related tasks, including simulation task design, task-aware scene generation, expert demonstration synthesis, and sim-to-real transfer, still demand substantial human effort. We present AnyTask, an automated framework that pairs massively parallel GPU simulation with foundation models to design diverse manipulation tasks and synthesize robot data. We introduce three AnyTask agents for generating expert demonstrations aiming to solve as many tasks as possible: 1) ViPR, a novel task and motion planning agent with VLM-in-the-loop Parallel Refinement; 2) ViPR-Eureka, a reinforcement learning agent with generated dense rewards and LLM-guided contact sampling; 3) ViPR-RL, a hybrid planning and learning approach that jointly produces high-quality demonstrations with only sparse rewards. We train behavior cloning policies on generated data, validate them in simulation, and deploy them directly on real robot hardware. The policies generalize to novel object poses, achieving 44% average success across a suite of real-world pick-and-place, drawer opening, contact-rich pushing, and long-horizon manipulation tasks. Our project website is at https://anytask.rai-inst.com .

cs.RO↗

arXiv2025

Sceniris: A Fast Procedural Scene Generation Framework

Synthetic 3D scenes are essential for developing Physical AI and generative models. Existing procedural generation methods often have low output throughput, creating a significant bottleneck in scaling up dataset creation. In this work, we introduce Sceniris, a highly efficient procedural scene generation framework for rapidly generating large-scale, collision-free scene variations. Sceniris also provides an optional robot reachability check, providing manipulation-feasible scenes for robot tasks. Sceniris is designed for maximum efficiency by addressing the primary performance limitations of the prior method, Scene Synthesizer. Leveraging batch sampling and faster collision checking in cuRobo, Sceniris achieves at least 234x speed-up over Scene Synthesizer. Sceniris also expands the object-wise spatial relationships available in prior work to support diverse scene requirements. Our code is available at https://github.com/rai-inst/sceniris

cs.RO↗

arXiv2025

Bridging the Regulatory Divide: Ensuring Safety and Equity in Wearable Health Technologies

As wearable health technologies have grown more sophisticated, the distinction between "wellness" and "medical" devices has become increasingly blurred. While some features undergo formal U.S. Food and Drug Administration (FDA) review, many over-the-counter tools operate in a regulatory grey zone, leveraging health-related data and outputs without clinical validation. Further complicating the issue is the widespread repurposing of wellness devices for medical uses, which can introduce safety risks beyond the reach of current oversight. Drawing on legal analysis, case studies, and ethical considerations, we propose an approach emphasizing distributed risk, patient-centered outcomes, and iterative reform. Without a more pluralistic and evolving framework, the promise of wearable health technology risks being undermined by growing inequities, misuse, and eroded public trust.

cs.CY↗

arXiv2025

Feature Selection via GANs (GANFS): Enhancing Machine Learning Models for DDoS Mitigation

Distributed Denial of Service (DDoS) attacks represent a persistent and evolving threat to modern networked systems, capable of causing large-scale service disruptions. The complexity of such attacks, often hidden within high-dimensional and redundant network traffic data, necessitates robust and intelligent feature selection techniques for effective detection. Traditional methods such as filter-based, wrapper-based, and embedded approaches, each offer strengths but struggle with scalability or adaptability in complex attack environments. In this study, we explore these existing techniques through a detailed comparative analysis and highlight their limitations when applied to large-scale DDoS detection tasks. Building upon these insights, we introduce a novel Generative Adversarial Network-based Feature Selection (GANFS) method that leverages adversarial learning dynamics to identify the most informative features. By training a GAN exclusively on attack traffic and employing a perturbation-based sensitivity analysis on the Discriminator, GANFS effectively ranks feature importance without relying on full supervision. Experimental evaluations using the CIC-DDoS2019 dataset demonstrate that GANFS not only improves the accuracy of downstream classifiers but also enhances computational efficiency by significantly reducing feature dimensionality. These results point to the potential of integrating generative learning models into cybersecurity pipelines to build more adaptive and scalable detection systems.

cs.CR↗

arXiv2024

VADER: Visual Affordance Detection and Error Recovery for Multi Robot Human Collaboration

Robots today can exploit the rich world knowledge of large language models to chain simple behavioral skills into long-horizon tasks. However, robots often get interrupted during long-horizon tasks due to primitive skill failures and dynamic environments. We propose VADER, a plan, execute, detect framework with seeking help as a new skill that enables robots to recover and complete long-horizon tasks with the help of humans or other robots. VADER leverages visual question answering (VQA) modules to detect visual affordances and recognize execution errors. It then generates prompts for a language model planner (LMP) which decides when to seek help from another robot or human to recover from errors in long-horizon task execution. We show the effectiveness of VADER with two long-horizon robotic tasks. Our pilot study showed that VADER is capable of performing complex long-horizon tasks by asking for help from another robot to clear a table. Our user study showed that VADER is capable of performing complex long-horizon tasks by asking for help from a human to clear a path. We gathered feedback from people (N=19) about the performance of the VADER performance vs. a robot that did not ask for help. https://google-vader.github.io/

cs.RO↗

arXiv2024

Automating Code Adaptation for MLOps -- A Benchmarking Study on LLMs

This paper explores the possibilities of the current generation of Large Language Models for incorporating Machine Learning Operations (MLOps) functionalities into ML training code bases. We evaluate the performance of OpenAI (gpt-3.5-turbo) and WizardCoder (open-source, 15B parameters) models on the automated accomplishment of various MLOps functionalities in different settings. We perform a benchmarking study that assesses the ability of these models to: (1) adapt existing code samples (Inlining) with component-specific MLOps functionality such as MLflow and Weights & Biases for experiment tracking, Optuna for hyperparameter optimization etc., and (2) perform the task of Translation from one component of an MLOps functionality to another, e.g., translating existing GitPython library based version control code to Data Version Control library based. We also propose three different approaches that involve teaching LLMs to comprehend the API documentation of the components as a reference while accomplishing the Translation tasks. In our evaluations, the gpt-3.5-turbo model significantly outperforms WizardCoder by achieving impressive Pass@3 accuracy in model optimization (55% compared to 0% by WizardCoder), experiment tracking (100%, compared to 62.5% by WizardCoder), model registration (92% compared to 42% by WizardCoder) and hyperparameter optimization (83% compared to 58% by WizardCoder) on average, in their best possible settings, showcasing its superior code adaptability performance in complex MLOps tasks.

cs.LG↗

arXiv2024

A State-of-the-practice Release-readiness Checklist for Generative AI-based Software Products

This paper investigates the complexities of integrating Large Language Models (LLMs) into software products, with a focus on the challenges encountered for determining their readiness for release. Our systematic review of grey literature identifies common challenges in deploying LLMs, ranging from pre-training and fine-tuning to user experience considerations. The study introduces a comprehensive checklist designed to guide practitioners in evaluating key release readiness aspects such as performance, monitoring, and deployment strategies, aiming to enhance the reliability and effectiveness of LLM-based applications in real-world settings.

cs.SE↗

arXiv2023

Hybrid Reinforcement Learning for Optimizing Pump Sustainability in Real-World Water Distribution Networks

This article addresses the pump-scheduling optimization problem to enhance real-time control of real-world water distribution networks (WDNs). Our primary objectives are to adhere to physical operational constraints while reducing energy consumption and operational costs. Traditional optimization techniques, such as evolution-based and genetic algorithms, often fall short due to their lack of convergence guarantees. Conversely, reinforcement learning (RL) stands out for its adaptability to uncertainties and reduced inference time, enabling real-time responsiveness. However, the effective implementation of RL is contingent on building accurate simulation models for WDNs, and prior applications have been limited by errors in simulation training data. These errors can potentially cause the RL agent to learn misleading patterns and actions and recommend suboptimal operational strategies. To overcome these challenges, we present an improved "hybrid RL" methodology. This method integrates the benefits of RL while anchoring it in historical data, which serves as a baseline to incrementally introduce optimal control recommendations. By leveraging operational data as a foundation for the agent's actions, we enhance the explainability of the agent's actions, foster more robust recommendations, and minimize error. Our findings demonstrate that the hybrid RL agent can significantly improve sustainability, operational efficiency, and dynamically adapt to emerging scenarios in real-world WDNs.

cs.AI↗

arXiv2023

Exploring Multilingual Text Data Distillation

With the rise of deep learning, large datasets and complex models have become common, requiring significant computing power. To address this, data distillation has emerged as a technique to quickly train models with lower memory and time requirements. However, data distillation on text-based datasets hasn't been explored much because of the challenges rising due to its discrete nature. Additionally, existing dataset distillation methods often struggle to generalize to new architectures. In the paper, we propose several data distillation techniques for multilingual text classification datasets using language-model-based learning methods. We conduct experiments to analyze their performance in terms of classification strength, and cross-architecture generalization. Furthermore, we investigate the language-specific fairness of the data summaries generated by these methods. Our approach builds upon existing techniques, enhancing cross-architecture generalization in the text data distillation domain.

cs.CL↗

arXiv2023

Green synthesis of silver nanoparticles using Curcuma longa flower extract and antibacterial activity

Silver nanoparticles (AgNP's) possess inherent biological potentials that have obliged an alternative, eco-friendly, sustainable approach to "Green Synthesis." In the present study, we synthesized Green Silver Nanoparticles (GAgNP's) using Curcuma longa L. (C. longa) flower extract as a reducing and capping agent. The synthesized GAgNP's were characterized using UV-Visible spectroscopy, X-ray diffraction (XRD), and High-resolution transmission electron microscopy (HR-TEM), which confirmed their homogeneity and physical characteristics. The GAgNP's were found to contain crystalline silver through XRD, and the particles were confirmed to be homogeneous and spherical with a size of approximately 5 nm, as evidenced by UV-Visible spectroscopy, XRD, and HR-TEM. In addition, the biological potential of GAgNP's was evaluated for their antibacterial activities. GAgNP's showed significant activity and formed different sizes of inhibition zones against all selected bacteria: Mycobacterium smegmatis (M. smegmatis) (26 mm), Mycobacterium phlei (M. phlei), and Staphylococcus aureus (S. aureus) (22 mm), Staphylococcus epidermidis (S. epidermidis) and Klebsiella pneumoniae (K. pneumoniae) (18 mm), and Escherichia coli (E. coli) (13 mm). The MIC value of GAgNP's was found to be between 625 ug/mL-39.06 ug/mL for different microbes tested. With further research, the green synthesis of GAgNP's using C. longa flower extracts could lead to the development of effective antibacterial treatments in the medical field.

q-bio.OT↗

arXiv2022

Using dynamic circles and squares to visualize spatio-temporal variation

Visualizations such as bar charts, scatter plots, and objects on geographical maps often convey critical information, including exact and relative numeric values, using shapes. The choice of shape and method of encoding information is often arbitrarily, or based on convention. However, past studies have shown that the human eye can be fooled by visual representations. The Ebbinghaus illusion demonstrates that the perceived relative sizes of shapes depends on their configuration, which in turn can affect judgements, especially in visualizations like proportional symbol maps. In this study we evaluate the effects of varying the type of shapes and metrics for encoding data in visual representations on a spatio-temporal map interface. We find that some combinations of shape and metric are more conducive to accurate human judgements than others, and provide recommendations for applying these findings in future visualization designs.

cs.HC↗

arXiv2022

Exploring Explainability Methods for Graph Neural Networks

With the growing use of deep learning methods, particularly graph neural networks, which encode intricate interconnectedness information, for a variety of real tasks, there is a necessity for explainability in such settings. In this paper, we demonstrate the applicability of popular explainability approaches on Graph Attention Networks (GAT) for a graph-based super-pixel image classification task. We assess the qualitative and quantitative performance of these techniques on three different datasets and describe our findings. The results shed a fresh light on the notion of explainability in GNNs, particularly GATs.

cs.LG↗

arXiv2022

Mid-Air Helicopter Delivery at Mars Using a Jetpack

Mid-Air Helicopter Delivery (MAHD) is a new Entry, Descent and Landing (EDL) architecture to enable in situ mobility for Mars science at lower cost than previous missions. It uses a jetpack to slow down a Mars Science Helicopter (MSH) after separation from the backshell, and reach aerodynamic conditions suitable for helicopter take-off in mid air. For given aeroshell dimensions, only MAHD's lander-free approach leaves enough room in the aeroshell to accommodate the largest rotor option for MSH. This drastically improves flight performance, notably allowing +150\% increased science payload mass. Compared to heritage EDL approaches, the simpler MAHD architecture is also likely to reduce cost, and enables access to more hazardous and higher-elevation terrains on Mars. This paper introduces a design for the MAHD system architecture and operations. We present a mechanical configuration that fits both MSH and the jetpack within the 2.65-m Mars heritage aeroshell, and a jetpack control architecture which fully leverages the available helicopter avionics. We discuss preliminary numerical models of the flow dynamics resulting from the interaction between the jets, the rotors and the side winds. We define a force-torque sensing architecture capable of handling the wind and trimming the rotors to prepare for safe take-off. Finally, we analyze the dynamic environment and closed-loop control simulation results to demonstrate the preliminary feasibility of MAHD.

cs.RO↗

arXiv2021

Orthosymplectic Cauchy identities

We give bijective proofs of orthosymplectic analogues of the Cauchy identity and dual Cauchy identity for orthosymplectic Schur functions. To do so, we present two insertion algorithms; these are orthosymplectic versions of Berele's symplectic insertion algorithms, which were used by Sundaram to give bijective proofs of Cauchy identities for symplectic Schur functions.

math.CO↗

arXiv2021

Trust as a Metric for Resiliency in Signed Social Networks

Recent technological advancements have resulted in a surge in online trading, raising severe concerns about theft and fraud, especially on platforms like Bitcoin OTC (over-the-counter), where users' identities remain anonymous. To mitigate the risk, it has become essential to capture the reputation of users based on their trade histories. The who-trusts-whom signed network of people has the capability to reflect the nature of such positive and negative relations between the users. It can be used to analyze linkage patterns, strength, and resiliency of such platforms. Due to the dynamic nature of trust between individuals, these trust networks are often vulnerable to link or node failures, making it critical to understand the stability of such systems. In this paper, we consider the problem of quantifying the resiliency of signed networks with the help of trustworthy community structures. We propose a metric for computing the Trustworthiness of a community structure. Using the trustworthiness scores of all communities structures, we generate a pipeline for assessing the resiliency of a signed network. We also show how these generated resiliency scores are concordant with the true nature of the network.

cs.SI↗

arXiv2021

Image Inpainting using Partial Convolution

Image Inpainting is one of the very popular tasks in the field of image processing with broad applications in computer vision. In various practical applications, images are often deteriorated by noise due to the presence of corrupted, lost, or undesirable information. There have been various restoration techniques used in the past with both classical and deep learning approaches for handling such issues. Some traditional methods include image restoration by filling gap pixels using the nearby known pixels or using the moving average over the same. The aim of this paper is to perform image inpainting using robust deep learning methods that use partial convolution layers.

cs.CV↗