arXiv subjects

Aditya Kumar

Publications and source records attributed to Aditya Kumar.

At least 19 recordsLinked to original sources

arXiv2026

6.5% of the Neuro-Symbolic Literature Can Be Reproduced from Its Published Artifacts, a Six-Stage Audit Framework and First Instantiation

We present a six-stage framework for auditing the reproducibility of scientific claims across a research literature within the computer science domain, and instantiate our framework for the neuro-symbolic AI (NSAI) subdomain. Instantiating the framework on the NSAI subdomain produced a multi-year audit. Stage one retrieved 5,497 records and removed 3,018 duplicates. Stage two screened the 2,479 unique records at title and abstract, identifying 1,365 self-identified NSAI records, then removed a further 61 at full text for off-topic, non-research, no-quantitative-evaluation, or inaccessible-full-text reasons. Stage three sought a verifiable public code artifact for each of the 1,304 eligible records and found none for 849, leaving 455 to enter the artifact inventory and bounded rerun of stages four and five. We fully or partially reproduced 85 studies, 6.52% of the eligible corpus and 18.68% of attempted reruns. We found that 321 attempted reruns were blocked by missing non- code artifacts and 42 by missing or unusable code repositories. These figures quantify a persistent reproducibility deficit that survives even nominal "code available" declarations, and signal the need for enforced, versioned, and permanently archived artifact bundles in future NSAI publications. We argue that empirical NSAI papers should be required at submission time to provide complete, versioned, and permanently archived artifact bundles.

cs.AI↗

arXiv2026

Dynamic Entanglement-Weighted Pruning for Quantum Federated Unlearning in Supply-Chain Risk Prediction

Federated deployments of variational quantum classifiers are attractive for cross-organisation risk prediction in supply chains, because raw data never leaves the client, yet data-protection regulations such as the GDPR grant clients a right to request that their contribution be removed from a trained model after the fact. Retraining a federated model from scratch to honour such a request is correct but wasteful, and it is not obvious which quantum circuit parameters actually carry a given client's influence. We introduce Entanglement-Weighted Pruning (EWP), an unlearning procedure for quantum federated learning that scores every trainable circuit parameter with the product of two signals: the diagonal entry of the quantum Fisher information matrix estimated on the target client's data via the parameter-shift rule, and a structural entanglement weight associated with the parameter's gate. Parameters with the lowest scores are pruned, optionally followed by a short fine-tuning pass on the retained clients. We implement the full pipeline in Qiskit for a four-qubit data-re-uploading ansatz trained with FedAvg across five simulated supply-chain-risk clients, and benchmark EWP against full retraining, fine-tuning alone, random pruning, Fisher-only pruning, and entanglement-only pruning, over three random seeds. EWP attains a mean post-unlearning accuracy statistically indistinguishable from the full-retraining oracle, while producing a lower forgetting score and requiring roughly 16 times less wall-clock time. Ablations over pruning threshold, client count, and non-IID strength show that combining the two signals is necessary, as entanglement-only and Fisher-only pruning each substantially degrade accuracy relative to EWP.

quant-ph↗

arXiv2026

Large bias-tunable magnetoresistance from spin-dependent interlayer hybridization in van der Waals antiferromagnet CrSBr-based heterostructures

We explore the large magnetoresistance (MR) in \ce{hBN}/few-layer-graphene/\ce{CrSBr}/few-layer-graphene heterostructures and reveal the mechanism behind its non-monotonic bias dependence. Using bias voltage and temperature as independent tuning knobs, we achieve MR up to \SI{350}{\percent} at \SI{20}{K}, characterized by symmetric M-shaped maxima around $\pm 0.5\,\mathrm{V}$. Continuous tuning of the magnetization angle $\theta$ via a hard-axis magnetic field shows that the barrier band-edge offset varies linearly with $\cos(\theta/2)$, a first-order signature of spin-dependent interlayer hybridization. This linear relationship rules out the Julli\`ere model and a spin-filter projection. We conclude that the magnetic-configuration-dependent band edge, rather than electrode spin polarization, dictates the large magnetoresistance in \ce{CrSBr} junctions.

cond-mat.mes-hall↗

arXiv2026

Finite-Time Analysis of Discounted Exponential-Utility Reinforcement Learning

Discounted exponential utility provides a principled criterion for risk-sensitive sequential decision-making, but its nonlinear structure complicates reinforcement learning. A recent work \citep{thoppe2026reinforcement} addressed this difficulty by introducing a Bellman-compatible surrogate and two model-free fixed-point algorithms for optimizing it over stationary policies. However, their main convergence results are asymptotic. In this work, we establish finite-time rates of $\tilde{O} (1/\sqrt{n})$ for the aforementioned two algorithms under asynchronous Markovian sampling, where $n$ is the iteration index and $\tilde{O}$ hides logarithmic expressions. Importantly, we employ parameter-free choices for the stepsize parameter to derive these rate results. For the algorithmically simpler one-timescale method, the main challenge is that its update equation is not directly aligned with the contraction geometry of its underlying power-law operator. We overcome this mismatch by exploiting the boundedness, monotonicity, and homogeneity of the operator to obtain a local pseudo-contraction property for the relative-error dynamics. We then use a Moreau-envelope-based Lyapunov function and Polyak--Ruppert averaging to obtain the stated convergence rate with parameter-free stepsizes. For the two-timescale method, the main challenge is to control a tracking error on the faster timescale. These results provide the first finite-time guarantees for model-free discounted exponential-utility reinforcement learning.

cs.LG↗

arXiv2026

Dynamic phase-field model for brittle fracture in grounded glaciers

Fracture and calving of glaciers are key contributors to ice-mass loss and sea-level rise, yet predictive modeling remains challenging. Fracture in grounded glaciers is driven by gravitational forces and is typically studied within the framework of quasi-static linear elastic fracture mechanics. In this work, we show that purely quasistatic brittle fracture simulations within the phase field fracture framework under fixed self-weight can become strongly overdriven after crevasse initiation, producing unphysical thickening of the diffusive crack band and diffuse damage patterns. This pathology arises because gravity drives a growing region ahead of the crack tip beyond the strength surface. To resolve this, we show that the post-nucleation propagation is fundamentally a dynamic instability rather than a quasistatic process and propose the use of dynamic formulations of fracture. We demonstrate that accounting for inertia results in sharp, localized cracks that propagate through the ice thickness. As a second objective, this paper introduces a new dynamic formulation of the phase-field fracture model of Kumar et al. (J. Mech. Phys. Solids 2018) in which elastic, inertial, and gravitational contributions are degraded consistently in fractured regions.

cond-mat.mtrl-sci↗

arXiv2026

Extending Detection Engineering to Digital Forensics: The Velociraptor Unified Detection-Forensics Methodology

Detection engineering and digital forensics have evolved in parallel rather than in partnership, leaving a gap between real-time alerting and forensic analysis. This paper develops a unified detection-forensics methodology using Velociraptor, where detection logic directly initiates targeted evidence acquisition at the point of detection. The contribution is threefold: (1) a four-stage methodology (baseline establishment, evidence correlation, attack chain analysis, and scenario labelling with confidence) that converts artefact knowledge into reusable and testable detection rules suitable for both post-incident triage and live monitoring; (2) a practical demonstration, using three Velociraptor BaseVQL log sources (/forensics/windows/prefetch, /forensics/windows/usn, and /windows/wmi) that practitioners can deploy today, showing that artefact-based detections enable scalable forensic triage without full disk acquisition; and (3) evidence that periodic artefact analysis offers continuous monitoring while substantially reducing data volume compared to conventional endpoint logging. Two case studies illustrate the approach: a Prefetch/USN baseline for triage when Windows Event Logs are cleared or unavailable, and a WMI persistence correlation supporting both triage and continuous monitoring through periodic artefact analysis.

cs.CR↗

arXiv2026

Concept Removal for Frontier Image Generative Models

Image generative models are trained on massive, largely uncurated internet-scale datasets that contain undesirable visual concepts. Efficiently removing such concepts from the model generations without degrading the quality of output images remains challenging. We introduce a novel concept removal method for frontier diffusion and image autoregressive models, such as SD3.5, Flux, and Infinity. Our intervention replaces the internal bottleneck layer present in all these modern models with a transcoder that is trained to replicate the original layer while structuring it into distinct activation features. This in-place substitution creates an integrated filter through which concept-specific signals can be selectively disabled while preserving the rest of the model's behavior. Since the intervention modifies the model backbone rather than attaching an external component, it remains persistent under white-box access. Empirically, the approach achieves state-of-the-art concept removal performance across modern diffusion and autoregressive models, maintains visual generation quality, provides robustness against adversarial prompts, and supports sequential removal of diverse concepts. This positions our method as a practical approach for concept removal in frontier image generative models.

cs.CV↗

arXiv2026

From Correlation to Causation in Lane Change Prediction for Automated Driving: A Causal Explanation Framework

Lane-change prediction is a central task in intelligent vehicles, where early maneuver anticipation can support safer decision-making. However, many existing approaches mainly learn statistical associations between observed driving variables and future maneuvers, while overlooking the causal dependencies among the input variables themselves. This limits interpretability, especially when physically related variables such as longitudinal gap, relative longitudinal velocity, and Time-To-Collision (TTC) are treated as independent flat inputs. This article presents a causal-inference-based framework for lane-change prediction and explanation. The proposed approach combines linguistic feature construction, expert-constrained causal discovery, deep structural causal modeling with Deep End-to-end Causal Inference (DECI), intervention-based effect analysis, refutation testing, and recursive causal-chain explanation. The objective is not only to predict the future maneuver, but also to identify candidate variables that directly contribute to the prediction, the upstream factors influencing them, and the causal chains through which these effects propagate. The framework achieves average F1-scores above 95% during the first three seconds before the lane-marking crossing event. Beyond prediction accuracy, the framework uses intervention-based effect analysis to distinguish influential from weakly influential variables under the learned causal structure. It further distinguishes candidate direct contributors from mediated effects and generates contrastive causal-chain explanations that clarify why the predicted maneuver is favored and why the alternative maneuvers are less supported. The main contribution is therefore a mechanism-aware lane-change prediction pipeline that moves beyond correlation-based classification toward more interpretable causal reasoning for maneuver prediction.

cs.LG↗

arXiv2026

Learning to Construct Practical Agentic Systems

Automated design and optimization of agentic LLM-based systems leads to sophisticated systems that substantially improve result quality over off-the-shelf agentic patterns. However, studies of fielded agentic systems show that production systems focus much more on issues such as simplicity, controllability, and predictability of inference costs. In this paper we propose principled approaches to designing and optimizing practical agentic systems. We describe an agent framework that enables designers to enforce modularity in agentic systems, by defining "pseudo-tools" that call LLMs recursively on a restricted context. Using this framework we hand-engineer agents for a diverse set of tasks, and show that relative to dynamically-planned workflows, hand-constructed fixed workflows are generally cheaper and more accurate. We then propose novel learning methods for the agentic components required by this framework, namely pseudo-tools and fixed workflows. These learning methods generally outperform hand-engineered agents. We also exploit the modularity of the framework to apply multi-objective optimization methods to jointly optimize cost and response quality and blend the results of multiple learning systems.

cs.LG↗

arXiv2026

SoK: After Decades of Web Tracker Detection, What's Next?

Web tracking is an omnipresent phenomenon in today's web, affecting users in their day-to-day lives. Filter lists and blockers were invented to detect trackers and to protect users. Due to limitations of said tools, researchers developed web tracker detectors to replace them. No review constructed a universal perspective and classification of web tracker detectors until now. Past reviews focused either on the field as a whole or on web tracking techniques. In this SoK paper, we present the most comprehensive meta-science study on web tracker detection by systematizing and synthesizing the available knowledge. We conduct a systematic review, resulting in 59 primary and 16 supplementary studies out of a corpus of 832 papers. Based on these findings we suggest a taxonomy, observe and evaluate trends, propose open research gaps, and recommendations with which we aim to lay the foundations for future web tracker detection research. In addition, we conduct a limited reproducibility study to assess the validity of past studies and highlight emerging problems in this field.

cs.CR↗

arXiv2026

Rational Mechanics of Material Strength in Brittle Solids

Material strength is a classical concept with renewed importance in fracture mechanics, particularly in crack nucleation in brittle solids. We formulate material strength in finite elasticity and examine its geometric, constitutive, and symmetry-theoretic foundations. Spatial covariance requires a strength function to depend on both stress and the corresponding strain measure, so that strength is governed by the pair (stress,strain), not stress alone, and only then can representations based on different stress measures be consistently related, with classical stress-based criteria recovered as a special case. We analyze covariance under spatial diffeomorphisms and relate formulations based on the first Piola--Kirchhoff, second Piola--Kirchhoff, and Cauchy stresses. For stress-based criteria, we define the strength hypersurface as a subset of the constitutively admissible stress manifold and study the associated safe domain. Under standard regularity assumptions and the requirement that sufficiently large stresses are inadmissible, the strength surface is a smooth compact hypersurface of this manifold. For isotropic solids, we show that the safe domain is star-shaped under a proportional-reduction hypothesis. We extend the formulation to anelastic brittle solids, showing that residual stresses and eigenstrains modify the strength surface through the material metric, and discuss anisotropic strength via material symmetry.

cond-mat.mtrl-sci↗

arXiv2026

Stable systolic inequalities via mod n covering

We introduce a mod $n$ covering based approach to stable systolic inequalities. The idea is to prescribe a cohomology class mod $n$ which forces the desired cup product or index to be nonzero, and then find a short integral lift of that class. The method is especially effective in rank two as we can compute the covering constant. As a curvature free application, we improve the stable two systolic bound for $S^2\times S^2$ to $2$. The same bound holds for every oriented four manifold with $b_2=2$. Under a positive scalar curvature lower bound, the mod $n$ covering method combined with a sharp cowaist inequality for line bundles gives stable two systolic bounds. This gives the sharp stable two systolic inequality for odd complex projective spaces and an $O(m\log m)$ bound for $(S^2)^m$ when scalar curvature is at least $2m$. For $S^2\times S^2$ one gets that every metric with scalar curvature at least $4$ has stable two systole at most $8\pi$.

math.DG↗

arXiv2026

Localizing and Correcting Errors for LLM-based Planners

Large language models (LLMs) have demonstrated strong reasoning capabilities on math and coding, but frequently fail on symbolic classical planning tasks. Our studies, as well as prior work, show that LLM-generated plans routinely violate domain constraints given in their instructions (e.g., walking through walls). To address this failure, we propose iteratively augmenting instructions with Localized In-Context Learning (L-ICL) demonstrations: targeted corrections for specific failing steps. Specifically, L-ICL identifies the first constraint violation in a trace and injects a minimal input-output example giving the correct behavior for the failing step. Our proposed technique of L-ICL is much effective than explicit instructions or traditional ICL, which adds complete problem-solving trajectories, and many other baselines. For example, on an 8x8 gridworld, L-ICL produces valid plans 89% of the time with only 60 training examples, compared to 59% for the best baseline, an increase of 30%. L-ICL also shows dramatic improvements in other domains (gridworld navigation, mazes, Sokoban, and BlocksWorld), and on several LLM architectures.

cs.AI↗

arXiv2026

Learning temporal embeddings from electronic health records of chronic kidney disease patients

We investigate whether temporal embedding models trained on longitudinal electronic health records can learn clinically meaningful representations without compromising predictive performance, and how architectural choices affect embedding quality. Model-guided medicine requires representations that capture disease dynamics while remaining transparent and task agnostic, whereas most clinical prediction models are optimised for a single task. Representation learning facilitates learning embeddings that generalise across downstream tasks, and recurrent architectures are well-suited for modelling temporal structure in observational clinical data. Using the MIMIC-IV dataset, we study patients with chronic kidney disease (CKD) and compare three recurrent architectures: a vanilla LSTM, an attention-augmented LSTM, and a time-aware LSTM (T-LSTM). All models are trained both as embedding models and as direct end-to-end predictors. Embedding quality is evaluated via CKD stage clustering and in-ICU mortality prediction. The T-LSTM produces more structured embeddings, achieving a lower Davies-Bouldin Index (DBI = 9.91) and higher CKD stage classification accuracy (0.74) than the vanilla LSTM (DBI = 15.85, accuracy = 0.63) and attention-augmented LSTM (DBI = 20.72, accuracy = 0.67). For in-ICU mortality prediction, embedding models consistently outperform end-to-end predictors, improving accuracy from 0.72-0.75 to 0.82-0.83, which indicates that learning embeddings as an intermediate step is more effective than direct end-to-end learning.

cs.LG↗

arXiv2026

Urysohn width and macroscopic scalar curvature

We show that the macroscopic version of Gromov's Urysohn width conjecture for scalar curvature is false in dimensions four and above. This is based on (1) a novel estimate on the codimension two Urysohn width of circle bundles over manifolds with large hypersphericity radius, and (2) a notion of ruling for Riemannian manifolds that yields circle bundles with total spaces admitting metrics of positive macroscopic scalar curvature. Along the way, we also show that Urysohn width is not continuous under Cheeger-Gromov collapsing limits. This article is a continuation of our study of metric invariants and scalar curvature for circle bundles over large Riemannian manifolds initiated in [KS25].

math.DG↗

arXiv2026

Temporal Fusion Nexus: A task-agnostic multi-modal embedding model for clinical narratives and irregular time series in post-kidney transplant care

We introduce Temporal Fusion Nexus (TFN), a multi-modal and task-agnostic embedding model to integrate irregular time series and unstructured clinical narratives. We analysed TFN in post-kidney transplant (KTx) care, with a retrospective cohort of 3382 patients, on three key outcomes: graft loss, graft rejection, and mortality. Compared to state-of-the-art model in post KTx care, TFN achieved higher performance for graft loss (AUC 0.96 vs. 0.94) and graft rejection (AUC 0.84 vs. 0.74). In mortality prediction, TFN yielded an AUC of 0.86. TFN outperformed unimodal baselines (approx 10% AUC improvement over time series only baseline, approx 5% AUC improvement over time series with static patient data). Integrating clinical text improved performance across all tasks. Disentanglement metrics confirmed robust and interpretable latent factors in the embedding space, and SHAP-based attributions confirmed alignment with clinical reasoning. TFN has potential application in clinical tasks beyond KTx, where heterogeneous data sources, irregular longitudinal data, and rich narrative documentation are available.

cs.LG↗

arXiv2025

Application of machine learning to predict food processing level using Open Food Facts

Ultra-processed foods are increasingly linked to health issues like obesity, cardiovascular disease, type 2 diabetes, and mental health disorders due to poor nutritional quality. This first-of-its-kind study at such a scale uses machine learning to classify food processing levels (NOVA) based on the Open Food Facts dataset of over 900,000 products. Models including LightGBM, Random Forest, and CatBoost were trained on nutrient concentration data. LightGBM performed best, achieving 80-85% accuracy across different nutrient panels and effectively distinguishing minimally from ultra-processed foods. Exploratory analysis revealed strong associations between higher NOVA classes and lower Nutri-Scores, indicating poorer nutritional quality. Products in NOVA 3 and 4 also had higher carbon footprints and lower Eco-Scores, suggesting greater environmental impact. Allergen analysis identified gluten and milk as common in ultra-processed items, posing risks to sensitive individuals. Categories like Cakes and Snacks were dominant in higher NOVA classes, which also had more additives, highlighting the role of ingredient modification. This study, leveraging the largest dataset of NOVA-labeled products, emphasizes the health, environmental, and allergenic implications of food processing and showcases machine learning's value in scalable classification. A user-friendly web tool is available for NOVA prediction using nutrient data: https://cosylab.iiitd.edu.in/foodlabel/.

q-bio.BM↗

arXiv2025

Why planar cracks fragment into echelon cracks

Predicting the path and shape of growing cracks is fundamental to understanding of fracture. Under out-of-plane shear loading, an initially planar crack may spontaneously fragment into multiple cracks, forming a striking echelon crack pattern. Explaining this crack morphogenesis in brittle materials has been a long-standing open problem essential to developing a complete theory of crack growth. Here, through comparison with classical experiments, we show that a strength-constrained minimization of the sum of elastic and surface energies explains echelon crack formation. Results are presented for both soft and hard materials, confirming the model's general applicability to any brittle material. As a corollary, we show that, contrary to prevailing views, a purely energetic minimization model is insufficient to predict the growth of large cracks. We identify two key non-dimensional parameters governing crack fragmentation and orientation, and demonstrate that these reconcile the various energy-based and stress-based empirical criteria proposed in the literature for crack path.

cond-mat.mtrl-sci↗