arXiv subjects

Siddharth Jain

Publications and source records attributed to Siddharth Jain.

At least 19 recordsLinked to original sources

arXiv2026

Sequential vs. Simultaneous Entanglement Swapping under Optimal Link-Layer Control

Connection-less, packet-switched quantum network architectures distribute entanglement across multi-hop paths through sequential entanglement swapping, in which each node acts on purely local state information. The architectural advantages over the connection-oriented alternative -- simultaneous SWAP-ASAP -- are compelling, but sequential swapping holds partial chains in intermediate buffers between successive swaps, exposing them to memory decoherence in a way simultaneous SWAP-ASAP avoids by design. We present a proof-of-principle study at fixed chain length $n = 4$ in which each elementary link is governed by a fixed reinforcement-learning policy optimizing the secret-key rate of the six-state protocol, leaving the network-layer protocol as the sole independent variable. Sweeping the network-layer memory coherence time $T_c^{\mathrm{ext}}$ over four orders of magnitude reveals a clear regime structure governed by the dimensionless ratio $T_c^{\mathrm{ext}}/\tau$, where $\tau$ is the per-link entanglement heralding latency. Simultaneous SWAP-ASAP delivers a constant rate across the full sweep. Sequential swapping, by contrast, collapses to zero end-to-end deliveries below $T_c^{\mathrm{ext}}/\tau = 25$, and begins recovering at $T_c^{\mathrm{ext}}/\tau = 50$. It remains limited by the simultaneous rate, which it saturates only at the relaxed end of the sweep. These results suggest that the connection-less penalty is a near-term phenomenon tied to present-day memory coherence rather than a fundamental property of sequential swapping.

quant-ph↗

arXiv2026

CUE-R: Beyond the Final Answer in Retrieval-Augmented Generation

As language models shift from single-shot answer generation toward multi-step reasoning that retrieves and consumes evidence mid-inference, evaluating the role of individual retrieved items becomes more important. Existing RAG evaluation typically targets final-answer quality, citation faithfulness, or answer-level attribution, but none of these directly targets the intervention-based, per-evidence-item utility view we study here. We introduce CUE-R, a lightweight intervention-based framework for measuring per-evidence-item operational utility in single-shot RAG using shallow observable retrieval-use traces. CUE-R perturbs individual evidence items via REMOVE, REPLACE, and DUPLICATE operators, then measures changes along three utility axes (correctness, proxy-based grounding faithfulness, and confidence error) plus a trace-divergence signal. We also outline an operational evidence-role taxonomy for interpreting intervention outcomes. Experiments on HotpotQA and 2WikiMultihopQA with Qwen-3 8B and GPT-5.2 reveal a consistent pattern: REMOVE and REPLACE substantially harm correctness and grounding while producing large trace shifts, whereas DUPLICATE is often answer-redundant yet not fully behaviorally neutral. A zero-retrieval control confirms that these effects arise from degradation of meaningful retrieval. A two-support ablation further shows that multi-hop evidence items can interact non-additively: removing both supports harms performance far more than either single removal. Our results suggest that answer-only evaluation misses important evidence effects and that intervention-based utility analysis is a practical complement for RAG evaluation.

cs.IR↗

arXiv2025

Simplifying Knowledge Transfer in Pretrained Models

Pretrained models are ubiquitous in the current deep learning landscape, offering strong results on a broad range of tasks. Recent works have shown that models differing in various design choices exhibit categorically diverse generalization behavior, resulting in one model grasping distinct data-specific insights unavailable to the other. In this paper, we propose to leverage large publicly available model repositories as an auxiliary source of model improvements. We introduce a data partitioning strategy where pretrained models autonomously adopt either the role of a student, seeking knowledge, or that of a teacher, imparting knowledge. Experiments across various tasks demonstrate the effectiveness of our proposed approach. In image classification, we improved the performance of ViT-B by approximately 1.4% through bidirectional knowledge transfer with ViT-T. For semantic segmentation, our method boosted all evaluation metrics by enabling knowledge transfer both within and across backbone architectures. In video saliency prediction, our approach achieved a new state-of-the-art. We further extend our approach to knowledge transfer between multiple models, leading to considerable performance improvements for all model participants.

cs.LG↗

arXiv2025

Minimalistic Video Saliency Prediction via Efficient Decoder & Spatio Temporal Action Cues

This paper introduces ViNet-S, a 36MB model based on the ViNet architecture with a U-Net design, featuring a lightweight decoder that significantly reduces model size and parameters without compromising performance. Additionally, ViNet-A (148MB) incorporates spatio-temporal action localization (STAL) features, differing from traditional video saliency models that use action classification backbones. Our studies show that an ensemble of ViNet-S and ViNet-A, by averaging predicted saliency maps, achieves state-of-the-art performance on three visual-only and six audio-visual saliency datasets, outperforming transformer-based models in both parameter efficiency and real-time performance, with ViNet-S reaching over 1000fps.

cs.CV↗

arXiv2023

The Next 700 ML-Enabled Compiler Optimizations

There is a growing interest in enhancing compiler optimizations with ML models, yet interactions between compilers and ML frameworks remain challenging. Some optimizations require tightly coupled models and compiler internals,raising issues with modularity, performance and framework independence. Practical deployment and transparency for the end-user are also important concerns. We propose ML-Compiler-Bridge to enable ML model development within a traditional Python framework while making end-to-end integration with an optimizing compiler possible and efficient. We evaluate it on both research and production use cases, for training and inference, over several optimization problems, multiple compilers and its versions, and gym infrastructures.

cs.PL↗

arXiv2023

Omitted Labels Induce Nontransitive Paradoxes in Causality

We explore "omitted label contexts," in which training data is limited to a subset of the possible labels. This setting is standard among specialized human experts or specific, focused studies. By studying Simpson's paradox, we observe that ``correct'' adjustments sometimes require non-exchangeable treatment and control groups. A generalization of Simpson's paradox leads us to study networks of conclusions drawn from different contexts, within which a paradox of nontransitivity arises. We prove that the space of possible nontransitive structures in these networks exactly corresponds to structures that form from aggregating ranked-choice votes.

cs.LG↗

arXiv2023

Jet substructure in neutral current deep inelastic e+p scattering at upcoming Electron-Ion Collider

Predictions are made for the jet substructure of one-jet events produced in electron-proton neutral current deep inelastic scattering at the future Electron-Ion Collider for exchanged four-momentum squared, $Q^2 > 125$ GeV$^2$. Data are simulated using Monte Carlo event generators PYTHIA 8.304 and RAPGAP 3.308 at the center of mass energies $\sqrt{s}$ = 63.2, 104.9 and 141 GeV. Jets and subjets are produced by longitudinally invariant $k_T$ cluster algorithm. The subjet multiplicity distributions and differential jet shapes are measured for different jet sizes and varying jet-resolution parameter. A comparison is presented between the $k_T$ and anti-$k_T$ cluster algorithms for the study of jets and subjets using the simulated data and the HERA data.

hep-ph↗

arXiv2023

RL4ReAl: Reinforcement Learning for Register Allocation

We aim to automate decades of research and experience in register allocation, leveraging machine learning. We tackle this problem by embedding a multi-agent reinforcement learning algorithm within LLVM, training it with the state of the art techniques. We formalize the constraints that precisely define the problem for a given instruction-set architecture, while ensuring that the generated code preserves semantic correctness. We also develop a gRPC based framework providing a modular and efficient compiler interface for training and inference. Our approach is architecture independent: we show experimental results targeting Intel x86 and ARM AArch64. Our results match or out-perform the heavily tuned, production-grade register allocators of LLVM.

cs.LG↗

arXiv2022

Fick-Jacobs description and first passage dynamics for diffusion in a channel under stochastic resetting

Transport of particles through channels is of paramount importance in physics, chemistry and surface science due to its broad real world applications. Much insights can be gained by observing the transition paths of a particle through a channel and collecting statistics on the lifetimes in the channel or the escape probabilities from the channel. In this paper, we consider the diffusive transport through a narrow conical channel of a Brownian particle subject to intermittent dynamics, namely, stochastic resetting. As such, resetting brings the particle back to a desired location from where it resumes its diffusive phase. To this end, we extend the Fick-Jacobs theory of channel-facilitated diffusive transport to resetting-induced transport. Exact expressions for the conditional mean first passage times, escape probabilities and the total average lifetime in the channel are obtained, and their behaviour as a function of the resetting rate are highlighted. It is shown that resetting can expedite the transport through the channel -- rigorous constraints for such conditions are then illustrated. Furthermore, we observe that a carefully chosen resetting rate can render the average lifetime of the particle inside the channel minimal. Interestingly, the optimal rate undergoes continuous and discontinuous transitions as some relevant system parameters are varied. The validity of our one-dimensional analysis and the corresponding theoretical predictions are supported by three-dimensional Brownian dynamics simulations. We thus believe that resetting can be useful to facilitate particle transport across biological membranes -- a phenomena that can spearhead further theoretical and experimental studies.

cond-mat.stat-mech↗

arXiv2021

Expert Graphs: Synthesizing New Expertise via Collaboration

Consider multiple experts with overlapping expertise working on a classification problem under uncertain input. What constitutes a consistent set of opinions? How can we predict the opinions of experts on missing sub-domains? In this paper, we define a framework of to analyze this problem, termed "expert graphs." In an expert graph, vertices represent classes and edges represent binary opinions on the topics of their vertices. We derive necessary conditions for expert graph validity and use them to create "synthetic experts" which describe opinions consistent with the observed opinions of other experts. We show this framework to be equivalent to the well-studied linear ordering polytope. We show our conditions are not sufficient for describing all expert graphs on cliques, but are sufficient for cycles.

cs.LG↗

arXiv2020

Robust Correction of Sampling Bias Using Cumulative Distribution Functions

Varying domains and biased datasets can lead to differences between the training and the target distributions, known as covariate shift. Current approaches for alleviating this often rely on estimating the ratio of training and target probability density functions. These techniques require parameter tuning and can be unstable across different datasets. We present a new method for handling covariate shift using the empirical cumulative distribution function estimates of the target distribution by a rigorous generalization of a recent idea proposed by Vapnik and Izmailov. Further, we show experimentally that our method is more robust in its predictions, is not reliant on parameter tuning and shows similar classification performance compared to the current state-of-the-art techniques on synthetic and real datasets.

stat.ML↗

arXiv2020

Coding for Optimized Writing Rate in DNA Storage

A method for encoding information in DNA sequences is described. The method is based on the precision-resolution framework, and is aimed to work in conjunction with a recently suggested terminator-free template independent DNA synthesis method. The suggested method optimizes the amount of information bits per synthesis time unit, namely, the writing rate. Additionally, the encoding scheme studied here takes into account the existence of multiple copies of the DNA sequence, which are independently distorted. Finally, quantizers for various run-length distributions are designed.

cs.IT↗

arXiv2020

What is the Value of Data? On Mathematical Methods for Data Quality Estimation

Data is one of the most important assets of the information age, and its societal impact is undisputed. Yet, rigorous methods of assessing the quality of data are lacking. In this paper, we propose a formal definition for the quality of a given dataset. We assess a dataset's quality by a quantity we call the expected diameter, which measures the expected disagreement between two randomly chosen hypotheses that explain it, and has recently found applications in active learning. We focus on Boolean hyperplanes, and utilize a collection of Fourier analytic, algebraic, and probabilistic methods to come up with theoretical guarantees and practical solutions for the computation of the expected diameter. We also study the behaviour of the expected diameter on algebraically structured datasets, conduct experiments that validate this notion of quality, and demonstrate the feasibility of our techniques.

cs.LG↗

arXiv2020

CodNN -- Robust Neural Networks From Coded Classification

Deep Neural Networks (DNNs) are a revolutionary force in the ongoing information revolution, and yet their intrinsic properties remain a mystery. In particular, it is widely known that DNNs are highly sensitive to noise, whether adversarial or random. This poses a fundamental challenge for hardware implementations of DNNs, and for their deployment in critical applications such as autonomous driving. In this paper we construct robust DNNs via error correcting codes. By our approach, either the data or internal layers of the DNN are coded with error correcting codes, and successful computation under noise is guaranteed. Since DNNs can be seen as a layered concatenation of classification tasks, our research begins with the core task of classifying noisy coded inputs, and progresses towards robust DNNs. We focus on binary data and linear codes. Our main result is that the prevalent parity code can guarantee robustness for a large family of DNNs, which includes the recently popularized binarized neural networks. Further, we show that the coded classification problem has a deep connection to Fourier analysis of Boolean functions. In contrast to existing solutions in the literature, our results do not rely on altering the training process of the DNN, and provide mathematically rigorous guarantees rather than experimental evidence.

cs.LG↗

arXiv2016

Duplication Distance to the Root for Binary Sequences

We study the tandem duplication distance between binary sequences and their roots. In other words, the quantity of interest is the number of tandem duplication operations of the form $\seq x = \seq a \seq b \seq c \to \seq y = \seq a \seq b \seq b \seq c$, where $\seq x$ and $\seq y$ are sequences and $\seq a$, $\seq b$, and $\seq c$ are their substrings, needed to generate a binary sequence of length $n$ starting from a square-free sequence from the set $\{0,1,01,10,010,101\}$. This problem is a restricted case of finding the duplication/deduplication distance between two sequences, defined as the minimum number of duplication and deduplication operations required to transform one sequence to the other. We consider both exact and approximate tandem duplications. For exact duplication, denoting the maximum distance to the root of a sequence of length $n$ by $f(n)$, we prove that $f(n)=Θ(n)$. For the case of approximate duplication, where a $β$-fraction of symbols may be duplicated incorrectly, we show that the maximum distance has a sharp transition from linear in $n$ to logarithmic at $β=1/2$. We also study the duplication distance to the root for sequences with a given root and for special classes of sequences, namely, the de Bruijn sequences, the Thue-Morse sequence, and the Fibbonaci words. The problem is motivated by genomic tandem duplication mutations and the smallest number of tandem duplication events required to generate a given biological sequence.

cs.IT↗

arXiv2016

Duplication-Correcting Codes for Data Storage in the DNA of Living Organisms

The ability to store data in the DNA of a living organism has applications in a variety of areas including synthetic biology and watermarking of patented genetically-modified organisms. Data stored in this medium is subject to errors arising from various mutations, such as point mutations, indels, and tandem duplication, which need to be corrected to maintain data integrity. In this paper, we provide error-correcting codes for errors caused by tandem duplications, which create a copy of a block of the sequence and insert it in a tandem manner, i.e., next to the original. In particular, we present two families of codes for correcting errors due to tandem-duplications of a fixed length, the first family can correct any number of errors while the second corrects a bounded number of errors. We also study codes for correcting tandem duplications of length up to a given constant $k$, where we are primarily focused on the cases of $k=2,3$. Finally, we provide a full classification of the sets of lengths allowed in tandem duplication that result in a unique root for all sequences.

cs.IT↗

arXiv2015

Capacity and Expressiveness of Genomic Tandem Duplication

The majority of the human genome consists of repeated sequences. An important type of repeated sequences common in the human genome are tandem repeats, where identical copies appear next to each other. For example, in the sequence $AGTC\underline{TGTG}C$, $TGTG$ is a tandem repeat, that may be generated from $AGTCTGC$ by a tandem duplication of length $2$. In this work, we investigate the possibility of generating a large number of sequences from a \textit{seed}, i.e.\ a small initial string, by tandem duplications of bounded length. We study the capacity of such a system, a notion that quantifies the system's generating power. Our results include \textit{exact capacity} values for certain tandem duplication string systems. In addition, motivated by the role of DNA sequences in expressing proteins via RNA and the genetic code, we define the notion of the \textit{expressiveness} of a tandem duplication system as the capability of expressing arbitrary substrings. We then \textit{completely} characterize the expressiveness of tandem duplication systems for general alphabet sizes and duplication lengths. In particular, based on a celebrated result by Axel Thue from 1906, presenting a construction for ternary square-free sequences, we show that for alphabets of size 4 or larger, bounded tandem duplication systems, regardless of the seed and the bound on duplication length, are not fully expressive, i.e. they cannot generate all strings even as substrings of other strings. Note that the alphabet of size 4 is of particular interest as it pertains to the genomic alphabet. Building on this result, we also show that these systems do not have full capacity. In general, our results illustrate that duplication lengths play a more significant role than the seed in generating a large number of sequences for these systems.

cs.IT↗

arXiv2014

On Match Lengths, Zero Entropy and Large Deviations - with Application to Sliding Window Lempel-Ziv Algorithm

The Sliding Window Lempel-Ziv (SWLZ) algorithm that makes use of recurrence times and match lengths has been studied from various perspectives in information theory literature. In this paper, we undertake a finer study of these quantities under two different scenarios, i) \emph{zero entropy} sources that are characterized by strong long-term memory, and ii) the processes with weak memory as described through various mixing conditions. For zero entropy sources, a general statement on match length is obtained. It is used in the proof of almost sure optimality of Fixed Shift Variant of Lempel-Ziv (FSLZ) and SWLZ algorithms given in literature. Through an example of stationary and ergodic processes generated by an irrational rotation we establish that for a window of size $n_w$, a compression ratio given by $O(\frac{\log n_w}{{n_w}^a})$ where $a$ depends on $n_w$ and approaches 1 as $n_w \rightarrow \infty$, is obtained under the application of FSLZ and SWLZ algorithms. Also, we give a general expression for the compression ratio for a class of stationary and ergodic processes with zero entropy. Next, we extend the study of Ornstein and Weiss on the asymptotic behavior of the \emph{normalized} version of recurrence times and establish the \emph{large deviation property} (LDP) for a class of mixing processes. Also, an estimator of entropy based on recurrence times is proposed for which large deviation principle is proved for sources satisfying similar mixing conditions.

cs.IT↗