arXiv subjects

Aditya Jain

Publications and source records attributed to Aditya Jain.

At least 19 recordsLinked to original sources

arXiv2026

Superspace invariants and 3-point correlators in 3d $\mathcal{N}=3,4$ SCFTs

We use the auxiliary polarization spinor formalism together with superspace techniques to construct a complete and minimal list of 3-point invariant structures in three-dimensional superconformal field theories (SCFTs) with $\mathcal{N}=3$ and $\mathcal{N}=4$ superconformal symmetry. The existence of non-abelian $R$-symmetry for $\mathcal{N}=3,4$ gives rise to novel invariant structures built from the antisymmetric invariant tensor. These invariants are used to enumerate the structural form of spinning 3-point correlators of general as well as conserved spinning superfield operators in 3d SCFTs. For conserved operators, we find that the $\mathcal{N}=3$ correlators are fixed upto one parity-even and one parity-odd structure, while $\mathcal{N}=4$ conserved correlators admit two parity-even structures and one parity-odd structure, with the second parity-even structure associated with mirror symmetry breaking.

hep-th↗

arXiv2026

S-matrices in the holomorphic modular bootstrap approach

We numerically determine the S-matrix by using connection formulae in the modular linear differential equation (MLDE) approach to the holomorphic modular bootstrap. We then determine exact formulae using the fact that entries in the $S$-matrix are integer entries in a cyclotomic extension of the field of rational numbers. This provides a method that is intrinsic to the MLDE setup and does not require inputs outside this framework. The method is illustrated with a selection of examples.

hep-th↗

arXiv2026

Stress Tests REVEAL Fragile Temporal and Visual Grounding in Video-Language Models

This work investigates a fundamental question: Do Video-Language Models (VidLMs) robustly account for video content, temporal sequence, and motion? Our investigation shows that, surprisingly, they often do not. We introduce REVEAL{}, a diagnostic benchmark that probes fundamental weaknesses of contemporary VidLMs through five controlled stress tests; assessing temporal expectation bias, reliance on language-only shortcuts, video sycophancy, camera motion sensitivity, and robustness to spatiotemporal occlusion. We test leading open- and closed-source VidLMs and find that these models confidently describe reversed scenes as forward, answer questions while neglecting video content, agree with false claims, struggle with basic camera motion, and fail to aggregate temporal information amidst simple spatiotemporal masking. Humans, on the other hand, succeed at these tasks with ease. Alongside our benchmark, we provide a data pipeline that automatically generates diagnostic examples for our stress tests, enabling broader and more scalable evaluation. We will release our benchmark and code to support future research.

cs.CV↗

arXiv2025

Enhancing Decoding Performance using Efficient Error Learning

Lowering the resource overhead needed to achieve fault-tolerant quantum computation is crucial to building scalable quantum computers. We show that adapting conventional maximum likelihood (ML) decoders to a small subset of efficiently learnable physical error characteristics can significantly improve the logical performance of a quantum error-correcting code. Specifically, we leverage error information obtained from efficient characterization methods based on Cycle Error Reconstruction (CER), which yields Pauli error rates on the $n$ qubits of an error-correcting code. Although the total number of Pauli error rates needed to describe a general noise process is exponentially large in $n$, we show that only a few of the largest few Pauli error rates are needed and that a heuristic technique can complete the Pauli error distribution for ML decoding from this restricted dataset. Using these techniques, we demonstrate significant performance improvements for decoding quantum codes under a variety of physically relevant error models. For instance, with CER data that constitute merely $1\%$ of the Pauli error rates in the system, we achieve a $10X$ gain in performance compared to the case where decoding is based solely on the fidelity of the underlying noise process. Our conclusions underscore the promise of recent error characterization methods for improving quantum error correction and lowering overheads.

quant-ph↗

arXiv2025

Open-Insect: Benchmarking Open-Set Recognition of Novel Species in Biodiversity Monitoring

Global biodiversity is declining at an unprecedented rate, yet little information is known about most species and how their populations are changing. Indeed, some 90% of Earth's species are estimated to be completely unknown. Machine learning has recently emerged as a promising tool to facilitate long-term, large-scale biodiversity monitoring, including algorithms for fine-grained classification of species from images. However, such algorithms typically are not designed to detect examples from categories unseen during training -- the problem of open-set recognition (OSR) -- limiting their applicability for highly diverse, poorly studied taxa such as insects. To address this gap, we introduce Open-Insect, a large-scale, fine-grained dataset to evaluate unknown species detection across different geographic regions with varying difficulty. We benchmark 38 OSR algorithms across three categories: post-hoc, training-time regularization, and training with auxiliary data, finding that simple post-hoc approaches remain a strong baseline. We also demonstrate how to leverage auxiliary data to improve species discovery in regions with limited data. Our results provide insights to guide the development of computer vision methods for biodiversity monitoring and species discovery.

cs.CV↗

arXiv2025

VLDBench Evaluating Multimodal Disinformation with Regulatory Alignment

Detecting disinformation that blends manipulated text and images has become increasingly challenging, as AI tools make synthetic content easy to generate and disseminate. While most existing AI safety benchmarks focus on single modality misinformation (i.e., false content shared without intent to deceive), intentional multimodal disinformation, such as propaganda or conspiracy theories that imitate credible news, remains largely unaddressed. We introduce the Vision-Language Disinformation Detection Benchmark (VLDBench), the first large-scale resource supporting both unimodal (text-only) and multimodal (text + image) disinformation detection. VLDBench comprises approximately 62,000 labeled text-image pairs across 13 categories, curated from 58 news outlets. Using a semi-automated pipeline followed by expert review, 22 domain experts invested over 500 hours to produce high-quality annotations with substantial inter-annotator agreement. Evaluations of state-of-the-art Large Language Models (LLMs) and Vision-Language Models (VLMs) on VLDBench show that incorporating visual cues improves detection accuracy by 5 to 35 percentage points over text-only models. VLDBench provides data and code for evaluation, fine-tuning, and robustness testing to support disinformation analysis. Developed in alignment with AI governance frameworks (e.g., the MIT AI Risk Repository), VLDBench offers a principled foundation for advancing trustworthy disinformation detection in multimodal media. Project: https://vectorinstitute.github.io/VLDBench/ Dataset: https://huggingface.co/datasets/vector-institute/VLDBench Code: https://github.com/VectorInstitute/VLDBench

cs.CL↗

arXiv2025

Who is Responsible? The Data, Models, Users or Regulations? A Comprehensive Survey on Responsible Generative AI for a Sustainable Future

Generative AI is rapidly moving from research to deployment, elevating the need for responsible development, evaluation, and governance. We conduct a PRISMA guided review of 232 studies (November 2022 - December 2025), spanning large language models, vision language models, diffusion models, and agentic pipelines. We make four contributions: (1) the first survey bridging governance principles, technical evaluation, and domain deployment across all four system types; (2) a ten-criterion rubric (C1-C10) scoring major AI safety benchmarks on risk-surface coverage, paired with a policy crosswalk mapping benchmarks to regulatory requirements; (3) twelve lifecycle KPIs, explainability guidance for foundation models, and a testbed catalogue; and (4) domain-specific analysis across healthcare, finance, education, arts, agriculture, and defense. Three findings emerge: benchmark coverage is dense for bias and toxicity but sparse for privacy, provenance, deepfakes, and system-level failures in agentic settings; evaluations remain largely static and task local, limiting audit portability; and inconsistent documentation complicates cross-release comparison. We outline a research agenda prioritizing adaptive multimodal evaluation, privacy and provenance testing, deepfake risk assessment, calibration reporting, versioned artifacts, and continuous monitoring. This survey offers a structured path to align generative AI evaluation with governance needs for safe and accountable deployment.

cs.CY↗

arXiv2024

Superspace invariants and correlators in 4d $\mathcal{N}=1$ superconformal field theories

Using polarization spinor methods in conjunction with the superspace formalism, we construct 3-point superconformal invariants that are used to determine the form of 3-point correlators of spinning superfield operators in $\mathcal{N}=1$ superconformal field theories (SCFTs) in 4-dimensions. We enumerate the structural form of various spinning 3-point correlators using these invariants and find additional constraints on their form when the operators are conserved supercurrents. For these purposes, we first construct the invariants and 3-point correlators in non-supersymmetric $4d$ CFTs which are then extended using superspace methods to $4d$ SCFTs.

hep-th↗

arXiv2024

Smallest quantum codes for amplitude damping noise

We describe the smallest quantum error correcting (QEC) code to correct for amplitude-damping (AD) noise, namely, a 3-qubit code that corrects all the single-qubit damping errors. We generalize this construction to a family of codes that correct AD noise up to any fixed order of the damping strength. We underpin the fundamental connection between the structure of our codes and the noise structure, via a relaxed form of the Knill-Laflamme conditions, different from existing formulations of approximate QEC conditions. Although the recovery procedure for this code is non-deterministic, our codes are optimal with respect to overheads and outperform existing codes to tackle AD noise in terms of entanglement fidelity. This formulation of probabilistic QEC further leads us to new family of quantum codes tailored to AD noise and also gives rise to a noise-adapted quantum Hamming bound for AD noise. Finally, we construct a set of universal logical gates for the 3-qubit code, thus providing a potential pathway to fault tolerance via this class of codes.

quant-ph↗

arXiv2024

Insect Identification in the Wild: The AMI Dataset

Insects represent half of all global biodiversity, yet many of the world's insects are disappearing, with severe implications for ecosystems and agriculture. Despite this crisis, data on insect diversity and abundance remain woefully inadequate, due to the scarcity of human experts and the lack of scalable tools for monitoring. Ecologists have started to adopt camera traps to record and study insects, and have proposed computer vision algorithms as an answer for scalable data processing. However, insect monitoring in the wild poses unique challenges that have not yet been addressed within computer vision, including the combination of long-tailed data, extremely similar classes, and significant distribution shifts. We provide the first large-scale machine learning benchmarks for fine-grained insect recognition, designed to match real-world tasks faced by ecologists. Our contributions include a curated dataset of images from citizen science platforms and museums, and an expert-annotated dataset drawn from automated camera traps across multiple continents, designed to test out-of-distribution generalization under field conditions. We train and evaluate a variety of baseline algorithms and introduce a combination of data augmentation techniques that enhance generalization across geographies and hardware setups.

cs.CV↗

arXiv2024

A machine learning pipeline for automated insect monitoring

Climate change and other anthropogenic factors have led to a catastrophic decline in insects, endangering both biodiversity and the ecosystem services on which human society depends. Data on insect abundance, however, remains woefully inadequate. Camera traps, conventionally used for monitoring terrestrial vertebrates, are now being modified for insects, especially moths. We describe a complete, open-source machine learning-based software pipeline for automated monitoring of moths via camera traps, including object detection, moth/non-moth classification, fine-grained identification of moth species, and tracking individuals. We believe that our tools, which are already in use across three continents, represent the future of massively scalable data collection in entomology.

cs.CV↗

arXiv2024

Agonist-Antagonist Pouch Motors: Bidirectional Soft Actuators Enhanced by Thermally Responsive Peltier Elements

In this study, we introduce a novel Mylar-based pouch motor design that leverages the reversible actuation capabilities of Peltier junctions to enable agonist-antagonist muscle mimicry in soft robotics. Addressing the limitations of traditional silicone-based materials, such as leakage and phase-change fluid degradation, our pouch motors filled with Novec 7000 provide a durable and leak-proof solution for geometric modeling. The integration of flexible Peltier junctions offers a significant advantage over conventional Joule heating methods by allowing active and reversible heating and cooling cycles. This innovation not only enhances the reliability and longevity of soft robotic applications but also broadens the scope of design possibilities, including the development of agonist-antagonist artificial muscles, grippers with can manipulate through flexion and extension, and an anchor-slip style simple crawler design. Our findings indicate that this approach could lead to more efficient, versatile, and durable robotic systems, marking a significant advancement in the field of soft robotics.

cs.RO↗

arXiv2023

Generation Z's Ability to Discriminate Between AI-generated and Human-Authored Text on Discord

The growing popularity of generative artificial intelligence (AI) chatbots such as ChatGPT is having transformative effects on social media. As the prevalence of AI-generated content grows, concerns have been raised regarding privacy and misinformation online. Among social media platforms, Discord enables AI integrations -- making their primarily "Generation Z" userbase particularly exposed to AI-generated content. We surveyed Generation Z aged individuals (n = 335) to evaluate their proficiency in discriminating between AI-generated and human-authored text on Discord. The investigation employed one-shot prompting of ChatGPT, disguised as a text message received on the Discord.com platform. We explore the influence of demographic factors on ability, as well as participants' familiarity with Discord and artificial intelligence technologies. We find that Generation Z individuals are unable to discern between AI and human-authored text (p = 0.011), and that those with lower self-reported familiarity with Discord demonstrated an improved ability in identifying human-authored compared to those with self-reported experience with AI (p << 0.0001). Our results suggest that there is a nuanced relationship between AI technology and popular modes of communication for Generation Z, contributing valuable insights into human-computer interactions, digital communication, and artificial intelligence literacy.

cs.HC↗

arXiv2023

Generator Assisted Mixture of Experts For Feature Acquisition in Batch

Given a set of observations, feature acquisition is about finding the subset of unobserved features which would enhance accuracy. Such problems have been explored in a sequential setting in prior work. Here, the model receives feedback from every new feature acquired and chooses to explore more features or to predict. However, sequential acquisition is not feasible in some settings where time is of the essence. We consider the problem of feature acquisition in batch, where the subset of features to be queried in batch is chosen based on the currently observed features, and then acquired as a batch, followed by prediction. We solve this problem using several technical innovations. First, we use a feature generator to draw a subset of the synthetic features for some examples, which reduces the cost of oracle queries. Second, to make the feature acquisition problem tractable for the large heterogeneous observed features, we partition the data into buckets, by borrowing tools from locality sensitive hashing and then train a mixture of experts model. Third, we design a tractable lower bound of the original objective. We use a greedy algorithm combined with model training to solve the underlying problem. Experiments with four datasets show that our approach outperforms these methods in terms of trade-off between accuracy and feature acquisition cost.

cs.LG↗

arXiv2023

Generative AI in Writing Research Papers: A New Type of Algorithmic Bias and Uncertainty in Scholarly Work

The use of artificial intelligence (AI) in research across all disciplines is becoming ubiquitous. However, this ubiquity is largely driven by hyperspecific AI models developed during scientific studies for accomplishing a well-defined, data-dense task. These AI models introduce apparent, human-recognizable biases because they are trained with finite, specific data sets and parameters. However, the efficacy of using large language models (LLMs) -- and LLM-powered generative AI tools, such as ChatGPT -- to assist the research process is currently indeterminate. These generative AI tools, trained on general and imperceptibly large datasets along with human feedback, present challenges in identifying and addressing biases. Furthermore, these models are susceptible to goal misgeneralization, hallucinations, and adversarial attacks such as red teaming prompts -- which can be unintentionally performed by human researchers, resulting in harmful outputs. These outputs are reinforced in research -- where an increasing number of individuals have begun to use generative AI to compose manuscripts. Efforts into AI interpretability lag behind development, and the implicit variations that occur when prompting and providing context to a chatbot introduce uncertainty and irreproducibility. We thereby find that incorporating generative AI in the process of writing research manuscripts introduces a new type of context-induced algorithmic bias and has unintended side effects that are largely detrimental to academia, knowledge production, and communicating research.

cs.CY↗

arXiv2023

Improved quantum error correction with randomized compiling

Current hardware for quantum computing suffers from high levels of noise, and so to achieve practical fault-tolerant quantum computing will require powerful and efficient methods to correct for errors in quantum circuits. Here, we explore the role and effectiveness of using noise tailoring techniques to improve the performance of error correcting codes. Noise tailoring methods such as randomized compiling (RC) convert complex coherent noise processes to effective stochastic noise. While it is known that this can be leveraged to design efficient diagnostic tools, we explore its impact on the performance of error correcting codes. Of particular interest is the important class of coherent errors, arising from control errors, where RC has the maximum effect -- converting these into purely stochastic errors. For these errors, we show here that RC delivers an improvement in performance of the concatenated Steane code by several orders of magnitude. We also show that below a threshold rotation angle, the gains in logical fidelity can be arbitrarily magnified by increasing the size of the codes. These results suggest that using randomized compiling can lead to a significant reduction in the resource overhead required to achieve fault tolerance.

quant-ph↗

arXiv2022

muBoost: An Effective Method for Solving Indic Multilingual Text Classification Problem

Text Classification is an integral part of many Natural Language Processing tasks such as sarcasm detection, sentiment analysis and many more such applications. Many e-commerce websites, social-media/entertainment platforms use such models to enhance user-experience to generate traffic and thus, revenue on their platforms. In this paper, we are presenting our solution to Multilingual Abusive Comment Identification Problem on Moj, an Indian video-sharing social networking service, powered by ShareChat. The problem dealt with detecting abusive comments, in 13 regional Indic languages such as Hindi, Telugu, Kannada etc., on the videos on Moj platform. Our solution utilizes the novel muBoost, an ensemble of CatBoost classifier models and Multilingual Representations for Indian Languages (MURIL) model, to produce SOTA performance on Indic text classification tasks. We were able to achieve a mean F1-score of 89.286 on the test data, an improvement over baseline MURIL model with a F1-score of 87.48.

cs.CL↗

arXiv2022

Machine Learning in Sports: A Case Study on Using Explainable Models for Predicting Outcomes of Volleyball Matches

Machine Learning has become an integral part of engineering design and decision making in several domains, including sports. Deep Neural Networks (DNNs) have been the state-of-the-art methods for predicting outcomes of professional sports events. However, apart from getting highly accurate predictions on these sports events outcomes, it is necessary to answer questions such as "Why did the model predict that Team A would win Match X against Team B?" DNNs are inherently black-box in nature. Therefore, it is required to provide high-quality interpretable, and understandable explanations for a model's prediction in sports. This paper explores a two-phased Explainable Artificial Intelligence(XAI) approach to predict outcomes of matches in the Brazilian volleyball League (SuperLiga). In the first phase, we directly use the interpretable rule-based ML models that provide a global understanding of the model's behaviors based on Boolean Rule Column Generation (BRCG; extracts simple AND-OR classification rules) and Logistic Regression (LogReg; allows to estimate the feature importance scores). In the second phase, we construct non-linear models such as Support Vector Machine (SVM) and Deep Neural Network (DNN) to obtain predictive performance on the volleyball matches' outcomes. We construct the "post-hoc" explanations for each data instance using ProtoDash, a method that finds prototypes in the training dataset that are most similar to the test instance, and SHAP, a method that estimates the contribution of each feature on the model's prediction. We evaluate the SHAP explanations using the faithfulness metric. Our results demonstrate the effectiveness of the explanations for the model's predictions.

cs.LG↗