arXiv subjects

Ritesh Kumar

Publications and source records attributed to Ritesh Kumar.

At least 19 recordsLinked to original sources

arXiv2026

Concordance-Optimal Continuous Music Emotion Estimation: Variance Decomposition, Dual-Branch Temporal Fusion, and Conformal Prediction Intervals

Continuous music emotion recognition (MER) estimates how Arousal and Valence evolve within a song, yet the field's standard metric, pooled frame-level correlation, is dominated by between-song differences and cannot distinguish a model that tracks emotional dynamics from one that only ranks songs by their average level. We decompose annotation variance into between- and within-song components and show that mean squared error (MSE) targets the former, while windowed Lin's concordance correlation coefficient (CCC = $r \cdot C_b$) targets the latter by penalising the variance compression MSE induces. A single weight $λ$ controls the trade-off: on 1,764 DEAM songs, raising $λ$ from 0 to 1 lowers pooled Arousal $r$ from 0.843 to 0.810 while raising mean within-song $r$ from 0.204 to 0.368. Building on this objective, we present MMAS (Multimodal Music Analysis System), an open-source model that predicts Arousal and Valence every 0.5 s from raw audio. It fuses frozen MERT-v1-95M embeddings with ten symbolic features from automatic transcription through a dual-branch temporal convolutional network with cross-modal attention. Because both branches are audio-derived, one model trains jointly on DEAM and PMEmo without domain conditioning. Under song-grouped 5-fold cross-validation (2,531 songs), pooled $r$ reaches 0.801/0.685 (Arousal/Valence) on DEAM and 0.742/0.746 on PMEmo. Against a ridge baseline on identical features, within-song Arousal $r$ roughly triples (0.344 vs. 0.125; $n = 353$ songs, Wilcoxon $p = 1.7 \times 10^{-21}$, $d = 0.56$). Complete transcription failure costs at most 0.04 pooled $r$. Split-conformal intervals attain 93.7% (Arousal) and 90.0% (Valence) marginal coverage at a 95% target, with the shortfall traced to within-song dependence. MMAS is released as a pip-installable package.

cs.SD↗

arXiv2026

Interpretable and Steerable Molecular Graph Generation using Concept Bottlenecks

Deep graph generators (DGGs) largely lack interpretability, which is a significant limitation for one of their most crucial applications, drug discovery, where interpretable and transparent generation could provide greater control over the desirability of generated molecules, thereby reducing trial-and-error in candidate selection. Common approaches to interpretability, such as disentanglement, cannot guarantee that the learned representations are human-interpretable. To address this, we propose incorporating concept bottlenecks into DGGs, allowing users to interpret, debug, and steer the generation process. We show that this consistently improves distributional fidelity (lower FCD), with only modest, non-systematic effects on validity and uniqueness. We further explore Concept Controller, an optimization-based steering approach that has proven to be effective for steerable image generation. We evaluate our approach on a one-shot molecular graph generation framework, though the underlying ideas can be transferred to sequential graph generation as well.

cs.LG↗

arXiv2026

GraphNOSE: A Graph Transformer in Olfaction

Predicting olfactory qualities from molecular structure is an open problem in chemoinformatics. Although linear models can link molecular features to odor descriptors, they often fail when extrapolating to novel chemical scaffolds, extreme molecular weights, or complex odor mixtures. To address this, we introduce GraphNOSE, an open-source graph transformer framework that predicts multi-label odor descriptors from simplified molecular-input line-entry system (SMILES) strings for single molecules and binary mixtures. By integrating positional and structural encodings within a transformer-based graph architecture, GraphNOSE achieves strong performance with six times fewer parameters than standard graph neural network (GNN) baseline while consistently outperforming linear models, molecular language model embeddings, molecular fingerprints, and baseline GNNs by an average area under the ROC curve (AUROC) margin of 4.52% (p < 0.01). GraphNOSE achieves an AUROC of 84% on out-of-distribution compounds (OODs). This exceeds the current state-of-the-art GNN for OOD in olfaction (Open-POM: 81%, p < 0.001), and identifies conditions under which linear models empirically fail. Finally, we apply XAI (explainable AI) methods to identify which substructures and molecular features drive odor predictions, yielding insights consistent with chemical intuition and grounded in the model's learned representations. Together, these results establish GraphNOSE as a scalable and interpretable architecture for olfactory prediction that generalizes to structurally distinct compounds underrepresented in current perceptual databases.

cs.LG↗

arXiv2026

text2ql: Multi-Target Natural Language Querying via a Language-Agnostic Intermediate Representation

Natural language interfaces to databases have traditionally suffered from three structural limitations: exclusive targeting of relational SQL, unconditional dependence on large language model (LLM) inference at query time, and absence of any runtime signal when generated queries are semantically incorrect. This paper presents text2ql, an open-source Python framework that addresses all three limitations through a language-agnostic Intermediate Representation (QueryIR) and a pluggable renderer architecture. A single seven-stage detection pipeline serves both SQL and GraphQL targets; a zero-LLM deterministic mode delivers 100% execution accuracy at a median latency of 3.2 ms with no API cost; and every generated query carries a runtime confidence score in [0.15, 0.97] computed from an additive signal model. Evaluated on 50-query random samples from the Spider and BIRD benchmarks (indicative results; full-set evaluation is planned), the LLM-backed mode achieves 62-70% exact match and 84-91% execution accuracy; the deterministic mode achieves 100% execution accuracy with zero parse errors across all 100 test cases. An ablation study isolates schema-aware prompting as the dominant accuracy lever, contributing +18.4 percentage points of exact-match gain over the schema-free baseline on both benchmarks. text2ql is publicly available at https://pypi.org/project/text2ql/ under the Apache 2.0 license.

cs.CL↗

arXiv2026

Leveraging Synthetic Data for Question Answering with Multilingual LLMs in the Agricultural Domain

Enabling farmers to access accurate agriculture-related information in their native languages in a timely manner is crucial for the success of the agriculture field. Publicly available general-purpose Large Language Models (LLMs) typically offer generic agriculture advisories, lacking precision in local and multilingual contexts. Our study addresses this limitation by generating multilingual (English, Hindi, Punjabi) synthetic datasets from agriculture-specific documents from India and fine-tuning LLMs for the task of question answering (QA). Evaluation on human-created datasets demonstrates significant improvements in factuality, relevance, and agricultural consensus for the fine-tuned LLMs compared to the baseline counterparts.

cs.CL↗

arXiv2026

One-Bit Distributed Mean Estimation with Unknown Variance

In this work, we study the problem of distributed mean estimation with 1-bit communication constraints when the variance is unknown. We focus on the setting where each user has access to one iid sample drawn from a distribution belonging to a location-scale family, and is limited to sending just a single bit of information to a central server whose goal is to estimate the mean. We propose simple non-adaptive and adaptive protocols and show that both achieve asymptotic normality. We derive bounds on the asymptotic (in the number of users) Mean Squared Error (MSE) achieved by these protocols. For a class of symmetric log-concave distributions, we derive matching lower bounds for the MSE of adaptive protocols, establishing the optimality of our scheme. Furthermore, we develop a lower bound on the MSE for non-adaptive protocols that applies to any symmetric strictly log-concave distribution, using a refined squared Hellinger distance analysis. Through this, we show that for many common distributions, including a subclass of the generalized Gaussian family, the asymptotic minimax MSE achieved by the best non-adaptive protocol is strictly larger than that achieved by our simple adaptive protocol. We also demonstrate that increasing the number of bits per user can only marginally reduce the asymptotic MSE of adaptive protocols. Our simulation results confirm a positive gap between the adaptive and non-adaptive settings, aligning with the theoretical bounds.

cs.IT↗

arXiv2026

Synergistic Interplay between Surface Polarons and Adsorbates for Photocatalytic Nitrogen Reduction on TiO$_2$(110)

Photocatalytic nitrogen reduction under ambient conditions represents a promising pathway toward sustainable ammonia production. However, the fundamental mechanisms, particularly the role of photogenerated charge carriers and their interactions with surface defects and adsorbates, remain elusive. Here, we employ density functional theory with Hubbard U corrections and hybrid functionals to demonstrate that the synergistic interactions between photogenerated electron polarons and point defects are essential for enabling nitrogen reduction on TiO$_2$(110). We reveal that water adsorption promotes polaron migration from subsurface to surface sites, while subsequent water dissociation stabilizes polarons near oxygen vacancies through proton coupled electron polaron transfer (PCEpT). This surface localization of polarons is critical for effective N$_2$ adsorption and activation. Our findings are consistent with previous experimental reports utilizing EPR that confirm the presence of reduced Ti species and STM, which shows the presence of water dimers on the surface. Moreover, the simultaneous interaction between polarons and reaction intermediates facilitates polaron transfer, thereby driving the completion of the nitrogen reduction reaction. Our findings elucidate the pivotal role of surface polarons in photocatalytic nitrogen fixation and provide mechanistic insights applicable to a broad range of oxide surfaces and interfaces capable of hosting small polarons, offering new design principles for efficient photocatalysts operating under ambient conditions.

cond-mat.mtrl-sci↗

arXiv2026

Trajectory-Informed Memory Generation for Self-Improving Agent Systems

LLM-powered agents face a persistent challenge: learning from their execution experiences to improve future performance. While agents can successfully complete many tasks, they often repeat inefficient patterns, fail to recover from similar errors, and miss opportunities to apply successful strategies from past executions. We present a novel framework for automatically extracting actionable learnings from agent execution trajectories and utilizing them to improve future performance through contextual memory retrieval. Our approach comprises four components: (1) a Trajectory Intelligence Extractor that performs semantic analysis of agent reasoning patterns, (2) a Decision Attribution Analyzer that identifies which decisions and reasoning steps led to failures, recoveries, or inefficiencies, (3) a Contextual Learning Generator that produces three types of guidance -- strategy tips from successful patterns, recovery tips from failure handling, and optimization tips from inefficient but successful executions, and (4) an Adaptive Memory Retrieval System that injects relevant learnings into agent prompts based on multi-dimensional similarity. Unlike existing memory systems that store generic conversational facts, our framework understands execution patterns, extracts structured learnings with provenance, and retrieves guidance tailored to specific task contexts. Evaluation on the AppWorld benchmark demonstrates consistent improvements, with up to 14.3 percentage point gains in scenario goal completion on held-out tasks and particularly strong benefits on complex tasks (28.5~pp scenario goal improvement, a 149\% relative increase).

cs.AI↗

arXiv2026

Fine-Tuning and Evaluating Conversational AI for Agricultural Advisory

Large Language Models show promise for agricultural advisory, yet vanilla models exhibit unsupported recommendations, generic advice lacking specific, actionable detail, and communication styles misaligned with smallholder farmer needs. In high stakes agricultural contexts, where recommendation accuracy has direct consequences for farmer outcomes, these limitations pose challenges for responsible deployment. We present a hybrid LLM architecture that decouples factual retrieval from conversational delivery: supervised fine-tuning with LoRA on expert-curated GOLDEN FACTS (atomic, verified units of agricultural knowledge) optimizes fact recall, while a separate stitching layer transforms retrieved facts into culturally appropriate, safety-aware responses. Our evaluation framework, DG-EVAL, performs atomic fact verification (measuring recall, precision, and contradiction detection) against expert-curated ground truth rather than Wikipedia or retrieved documents. Experiments across multiple model configurations on crops and queries from Bihar, India show that fine-tuning on curated data substantially improves fact recall and F1, while maintaining high relevance. Using a fine-tuned smaller model achieves comparable or better factual quality at a fraction of the cost of frontier models. A stitching layer further improves safety subscores while maintaining high conversational quality. We release the farmerchat-prompts library to enable reproducible development of domain-specific agricultural AI.

cs.CL↗

arXiv2026

VOILA: Value-of-Information Guided Fidelity Selection for Cost-Aware Multimodal Question Answering

Despite significant costs from retrieving and processing high-fidelity visual inputs, most multimodal vision-language systems operate at fixed fidelity levels. We introduce VOILA, a framework for Value-Of-Information-driven adaptive fidelity selection in Visual Question Answering (VQA) that optimizes what information to retrieve before model execution. Given a query, VOILA uses a two-stage pipeline: a gradient-boosted regressor estimates correctness likelihood at each fidelity from question features alone, then an isotonic calibrator refines these probabilities for reliable decision-making. The system selects the minimum-cost fidelity maximizing expected utility given predicted accuracy and retrieval costs. We evaluate VOILA across three deployment scenarios using five datasets (VQA-v2, GQA, TextVQA, LoCoMo, FloodNet) and six Vision-Language Models (VLMs) with 7B-235B parameters. VOILA consistently achieves 50-60% cost reductions while retaining 90-95% of full-resolution accuracy across diverse query types and model architectures, demonstrating that pre-retrieval fidelity selection is vital to optimize multimodal inference under resource constraints.

cs.CV↗

arXiv2025

Selective Amplification of the Topological Hall Signal in Cr$_2$Te$_3$: The Role of Molecular Exchange Coupling

Layered magnetic transition-metal chalcogenides (TMCs) are a focal point of research, revealing a variety of intriguing magnetic and topological ground states. Within this family of TMCs, chromium telluride has garnered significant attention because of its excellent tunability in magnetic response, owing to the presence of competing magnetic exchange interactions. We here demonstrate the manipulation of magnetic anisotropy in ultra-thin Cr$_2$Te$_3$ films through growth engineering leading to a controlled transition from in-plane to out-of-plane orientation with an intermediate non-coplanar magnetic ground phase characterized by a topological Hall effect. Moreover, interfacing these films with Vanadyl phthalocyanine (VOPc) molecules prominently enhances the non-coplanar magnetic phase, attributing its presence to the competing interfacial magnetic exchange interactions over the spin-orbit-driven interfacial effects. These findings pave the way for the realization of novel topological spintronic devices through interface-modulated exchange coupling.

cond-mat.mtrl-sci↗

arXiv2025

Unravelling magnetic vortex-like excitations through rapid thermal quenching in low-carbon steel

Steel, traditionally valued for its structural strength, emerges in this study as a remarkable material for exploring novel magnetic phenomena. We investigate how common processing techniques-thermal treatments and mechanical strain-significantly affect the magnetic properties of low-carbon steels (0.05 percent by weight). Our findings show that slow annealing enlarges the grain size, enhancing magnetic susceptibility, while rapid quenching reduces grain size, resulting in a decreased magnetic response. Quenching low-carbon steel produces significant increase in the fraction of high-angle grain boundaries and a rapid spatial variation of local magnetic anisotropy between grains, a feature which is unachievable with mechanical straining even up to the material's ultimate tensile strength. Tensile-straining of low-carbon steel enhances magnetic susceptibility through altered magnetic anisotropy, contrary to the observed decrease of susceptibility in quenched low-carbon steel. Magnetic force microscopy and micromagnetic modelling of our data reveal that, the reduced magnetic susceptibility in quenched steel is a result of the presence of intriguing magnetic excitations akin to magnetic vortices. These localized structures act as strong magnetic domain wall pinning centres, causing the observed decrease in magnetic susceptibility in these quenched low-carbon steels. Beyond its established structural utility, low-carbon steel combines mechanical stability with favourable magnetic properties, positioning it as a strong platform for magnetic device applications.

cond-mat.mtrl-sci↗

arXiv2025

Low Temperature Two Fluid State in SmB6

Comprehensive study using DC transport, specific heat, magnetization, and two-coil mutual inductance measurements unveils an understanding of three temperature regimes in SmB$_6$: (i) $T \geq T^{*}$ ($\sim66$K), (ii) $T_g$ ($\sim40$ K) $\leq T < T^{*}$, and (iii) $T < T_g$. Onset of Kondo breakdown below $T^{*}$ releases disorder-driven magnetic fluctuations, which splits the bulk ($\sim116$K) and surface Kondo temperature ($T_k^{s} \approx 7$ K). Below $T_g$, as magnetic fluctuations subside, surface Kondo screening revives, stabilizing the topological surface state and generating an in-gap feature ($\sim2.2$ meV) across which Dirac-like carriers are excited. Nyquist impedance analysis reveals a crossover from purely capacitive to capacitive-inductive behavior, signalling a disorder-driven two-fluid phase of heavy quasiparticles and light, high-mobility carriers below $T_g$. We identify a characteristic length scale, $L_{ν_0}(T)$, associated with the high-mobility phase, exhibiting an almost divergent trend below $T_k^{s}$. These findings underscore the complex nature of the surface conducting state in SmB$_6$.

cond-mat.str-el↗

arXiv2025

Convolutional Neural Network Optimization for Beehive Classification Using Bioacoustic Signals

The behavior of honeybees is an important ecological phenomenon not only in terms of honey and beeswax production but also due to the proliferation of flora and fauna around it. The best way to study this significant phenomenon is by non-invasive monitoring of beehives using the sounds produced by various body movements that give out audio signals which can be exploited for various predictions related to the objectives mentioned above. This study investigates the application of Convolutional Neural Networks to classify and monitor different hive states with the help of joint time and frequency image representations such as Spectrogram, Mel-Spectrogram, Smoothed-Spectrogram, and Cochleagram. Our findings indicate that the Cochleagram outperformed all the other representations, achieving an accuracy of 98.31% on unseen data. Furthermore, we employed various strategies including pruning, quantization, and knowledge distillation to optimize the network and prevent any potential issues with model size. With these optimizations, the network size was lowered by 91.8% and the inference time was accelerated by 66%, increasing its suitability for real-time applications. Thus our study emphasizes the significance of using optimization approaches to minimize model size, avoid deployment problems, and expedite inference for real-time application as well as the selection of an appropriate time-frequency representation for optimal performance.

cs.SD↗

arXiv2025

An Explainable AI based approach for Monitoring Animal Health

Monitoring cattle health and optimizing yield are key challenges faced by dairy farmers due to difficulties in tracking all animals on the farm. This work aims to showcase modern data-driven farming practices based on explainable machine learning(ML) methods that explain the activity and behaviour of dairy cattle (cows). Continuous data collection of 3-axis accelerometer sensors and usage of robust ML methodologies and algorithms, provide farmers and researchers with actionable information on cattle activity, allowing farmers to make informed decisions and incorporate sustainable practices. This study utilizes Bluetooth-based Internet of Things (IoT) devices and 4G networks for seamless data transmission, immediate analysis, inference generation, and explains the models performance with explainability frameworks. Special emphasis is put on the pre-processing of the accelerometers time series data, including the extraction of statistical characteristics, signal processing techniques, and lag-based features using the sliding window technique. Various hyperparameter-optimized ML models are evaluated across varying window lengths for activity classification. The k-nearest neighbour Classifier achieved the best performance, with AUC of mean 0.98 and standard deviation of 0.0026 on the training set and 0.99 on testing set). In order to ensure transparency, Explainable AI based frameworks such as SHAP is used to interpret feature importance that can be understood and used by practitioners. A detailed comparison of the important features, along with the stability analysis of selected features, supports development of explainable and practical ML models for sustainable livestock management.

cs.LG↗

arXiv2025

FLOW-BENCH: Towards Conversational Generation of Enterprise Workflows

Business process automation (BPA) that leverages Large Language Models (LLMs) to convert natural language (NL) instructions into structured business process artifacts is becoming a hot research topic. This paper makes two technical contributions -- (i) FLOW-BENCH, a high quality dataset of paired natural language instructions and structured business process definitions to evaluate NL-based BPA tools, and support bourgeoning research in this area, and (ii) FLOW-GEN, our approach to utilize LLMs to translate natural language into an intermediate representation with Python syntax that facilitates final conversion into widely adopted business process definition languages, such as BPMN and DMN. We bootstrap FLOW-BENCH by demonstrating how it can be used to evaluate the components of FLOW-GEN across eight LLMs of varying sizes. We hope that FLOW-GEN and FLOW-BENCH catalyze further research in BPA making it more accessible to novice and expert users.

cs.AI↗

arXiv2025

Navigating the Fragrance space Via Graph Generative Models And Predicting Odors

We explore a suite of generative modelling techniques to efficiently navigate and explore the complex landscapes of odor and the broader chemical space. Unlike traditional approaches, we not only generate molecules but also predict the odor likeliness with ROC AUC score of 0.97 and assign probable odor labels. We correlate odor likeliness with physicochemical features of molecules using machine learning techniques and leverage SHAP (SHapley Additive exPlanations) to demonstrate the interpretability of the function. The whole process involves four key stages: molecule generation, stringent sanitization checks for molecular validity, fragrance likeliness screening and odor prediction of the generated molecules. By making our code and trained models publicly accessible, we aim to facilitate broader adoption of our research across applications in fragrance discovery and olfactory research.

cs.LG↗

arXiv2024

From Phytochemicals to Recipes: Health Indications and Culinary Uses of Herbs and Spices

Herbs and spices each contain about 3000 phytochemicals on average and there is much traditional knowledge on their health benefits. However, there is a lack of systematic study to understand the relationship among herbs and spices, their phytochemical constituents, their potential health benefits, and their usage in regional cuisines. Here we use a network-based approach to elucidate established relationships and predict novel associations between the phytochemicals present in herbs and spices with health indications. Our top 100 inferred indication-phytochemical relationships rediscover 40% known relationships and 20% that have been inferred via gene-chemical interactions with high confidence. The remaining 40% are hypotheses generated in a principled way for further experimental investigations. We also develop an algorithm to find the minimum set of spices needed to cover a target group of health conditions. Drawing on spice usage patterns in several regional Indian cuisines, and a copy-mutate model for regional cuisine evolution, we characterize the spectrum of health conditions covered by existing regional cuisines. The spectrum of health conditions can expand through the nationalization/globalization of culinary practice.

physics.soc-ph↗