arXiv subjects

Vikas Kumar

Publications and source records attributed to Vikas Kumar.

At least 19 recordsLinked to original sources

arXiv2026

Reexamining zero-shot summarization: Empirical investigation of trustworthiness of LLM-summarizers

Zero-shot summarization using Large Language Models (LLMs) has significantly advanced the abstractive summarization task by producing coherent and fluent summaries. However, underlying stochasticity of the large language models raises concerns about the stability and trustworthiness of the LLM-generated summaries. This issue has become increasingly important due to proliferation of LLM-generated summaries in educational settings, where students and researchers summarize complex academic materials in zero-shot manner. We propose a novel two-level diagnostic protocol for benchmarking LLM-summarizers based on the stability of the generated summaries. At the lower level, document-level stability analysis is performed over multiple LLM-summaries generated under controlled environment, and the stability coefficient is computed. Each generated summary is scored for semantic and factual alignment with the original document, enabling estimation of stability along more than one dimensions. At the next level, observations from a stratified sample of documents drawn from the corpus are consolidated to estimate the stability index of the LLM-summarizer, which is the proxy for its trustworthiness. Our empirical investigation of three LLM-summarizers across three genres of documents reveals statistically significant differences in the generation-level variability among LLMs across summary evaluation metrics. This study advances the LLM-summarization research by evidential recognition of the stability problem in LLM-summaries and motivates further research towards development of robust, reliable and trustworthy LLM-summarizers.

cs.AI↗

arXiv2026

Simple Features and Honest Calibration for Ambivalence and Hesitancy Recognition in Video

We address ambivalence and hesitancy (A/H) recognition in the ABAW 2026 BAH Challenge: given a short interview video, predict whether the person shows signs of A/H. Our system combines affect-specialised text, audio, and visual representations with a small set of readable linguistic hesitation cues, fused by a reliability gate we call Affective Marker Fusion (AMF), and finished with a simple AP-weighted ensemble at a fixed decision threshold. We also introduce \emph{ASR-erased time}: speech recognisers delete fillers and hesitation pauses from the transcript, but the chunk timestamps keep the time those events took, and sixteen features built from these gaps form the strongest and most independent non-verbal channel we measured (AP $0.718$, correlation $0.11$--$0.36$ with all other members). Across controlled experiments we find three things: cross-modal conflict design does not reliably help on BAH; language is by far the strongest channel while affect-specialised audio is a useful second; and calibration matters more than architecture. Fitting ensemble weights and a threshold on the small validation split overfits: it scores $0.741$ macro-F1 on validation but only $0.690$ on the untouched test set. AP-weighting at a fixed threshold instead reaches $\mathbf{0.731}$ on test.

cs.CV↗

arXiv2026

PRecG: Legal Precedent Retrieval with Graph Neural Networks and Rhetorical Role Segmentation

Legal precedent retrieval is a fundamental task in legal case preparation, planning, litigation strategy, and legal research. Current approaches for automatic precedent retrieval map legal documents to a low-dimensional semantic space and compute similarity based on the proximity of their representations. These approaches treat legal documents as monolithic texts, ignoring the rhetorical organization of the legal technicalities. Ergo, they overlook nuanced legal meanings and fail to distinguish the contextual significance of legal entities and concepts that vary based on their rhetorical roles within the document. To address this insufficiency, we propose the PRecG pipeline that computes the similarity between pairs of legal judgments by hierarchically learning their representations. The process begins by decomposing each document into distinct semantic units (segments) based on the rhetorical roles of sentences. For each rhetorical segment, a knowledge graph is constructed to capture the legal entities and their relationships within the segment. Contextual representations of the entities are then learned and aggregated to derive segment-level embeddings. These embeddings are further integrated to produce a unified document-level representation, and finally, the semantic similarity between a pair of documents is computed. We validate the performance of the proposed approach through extensive experiments on a benchmark Indian legal dataset, comparing it against state-of-the-art baselines to demonstrate its effectiveness.

cs.CL↗

arXiv2026

DReX: An Explainable Deep Learning-based Multimodal Recommendation Framework

Multimodal recommender systems leverage diverse data sources, such as user interactions, content features, and contextual information, to address challenges like cold-start and data sparsity. However, existing methods often suffer from one or more key limitations: processing different modalities in isolation, requiring complete multimodal data for each interaction during training, or independent learning of user and item representations. These factors contribute to increased complexity and potential misalignment between user and item embeddings. To address these challenges, we propose DReX, a unified multimodal recommendation framework that incrementally refines user and item representations by leveraging interaction-level features from multimodal feedback. Our model employs gated recurrent units to selectively integrate these fine-grained features into global representations. This incremental update mechanism provides three key advantages: (1) simultaneous modeling of both nuanced interaction details and broader preference patterns, (2) eliminates the need for separate user and item feature extraction processes, leading to enhanced alignment in their learned representation, and (3) inherent robustness to varying or missing modalities. We evaluate the performance of the proposed approach on three real-world datasets containing reviews and ratings as interaction modalities. By considering review text as a modality, our approach automatically generates interpretable keyword profiles for both users and items, which supplement the recommendation process with interpretable preference indicators. Experiment results demonstrate that our approach outperforms state-of-the-art methods across all evaluated datasets.

cs.IR↗

arXiv2026

GrIT: Group Informed Transformer for Sequential Recommendation

Sequential recommender systems aim to predict a user's future interests by extracting temporal patterns from their behavioral history. Existing approaches typically employ transformer-based architectures to process long sequences of user interactions, capturing preference shifts by modeling temporal relationships between items. However, these methods often overlook the influence of group-level features that capture the collective behavior of similar users. We hypothesize that explicitly modeling temporally evolving group features alongside individual user histories can significantly enhance next-item recommendation. Our approach introduces latent group representations, where each user's affiliation to these groups is modeled through learnable, time-varying membership weights. The membership weights at each timestep are computed by modeling shifts in user preferences through their interaction history, where we incorporate both short-term and long-term user preferences. We extract a set of statistical features that capture the dynamics of user behavior and further refine them through a series of transformations to produce the final drift-aware membership weights. A group-based representation is derived by weighting latent group embeddings with the learned membership scores. This representation is integrated with the user's sequential representation within the transformer block to jointly capture personal and group-level temporal dynamics, producing richer embeddings that lead to more accurate, context-aware recommendations. We validate the effectiveness of our approach through extensive experiments on five benchmark datasets, where it consistently outperforms state-of-the-art sequential recommendation methods.

cs.IR↗

arXiv2026

Data-Efficient Curation for Multimodal Reasoning under Fixed Training Protocols

We study data curation for multimodal reasoning in a fixed-protocol fine-tuning regime, where the base model, optimizer, training schedule, and evaluation pipeline are held constant and the main degree of freedom is the training data. Using the NeurIPS 2025 Data Curation for Vision--Language Reasoning (DCVLR) challenge as a controlled testbed, we analyze how source-dataset alignment, model-relative difficulty, dataset size, diversity heuristics, and rewritten synthetic mixtures affect downstream reasoning accuracy. Among the tested interventions, difficulty filtering on an aligned source corpus provides the strongest gains at matched scale. The effect is not explained only by LiveXivTQA weighting: a per-benchmark decomposition shows that much of the improvement over random sampling comes from OlympiadBench, the largest non-LiveXivTQA benchmark. Qwen-derived difficulty scores also transfer to some additional model families, though the benefit is architecture-dependent. In contrast, increasing dataset size beyond roughly 1k aligned examples mainly reduces run-to-run variance under the fixed recipe, while the diversity and rewritten CoSyn mixtures we tested do not improve over the difficulty-filtered baseline. These results provide a scoped empirical recipe for data-constrained multimodal reasoning fine-tuning, rather than a universal claim about data selection across all training regimes.

cs.AI↗

arXiv2025

AugAbEx: Bridging Abstractive and Extractive Legal Summarization

Automatic summarization of legal judgments liberates law professionals from heavy cognitive burden due to the complexity of the language, context-sensitive legal jargon, and the length of the document. Caveats of abstractive summarization for legal documents, revealed in recent studies, have impelled the development of newhybrid and extractive summarization methods, along with datasets in the legal domain. We propose an efficient and elegant pipeline (AugAbEx) to repurpose an existing large case summarization dataset with human-written gold-standard summaries by augmenting it with silver standard extractive summaries ensconcing experts' opinion. Availability of silver summaries along with gold standard human-written summaries will bolster development and evaluation of novel hybrid and extractive case-summarization algorithms in the legal domain. We thoroughly scrutinize the augmented extractive summaries in structural, lexical, and semantic dimensions, within a domain specific framework, to ensure quality. Extensive experiments and statistical test on seven English legal case-summarization datasets demonstrate that the silver standard extractive summaries produced by the proposed pipeline score well across all evaluation dimensions. Comparison of AugAbEx with two baselines and three current state-of-the-art methods reveals that it outperforms all competing methods, and the quality of the summaries produced by the proposed pipeline is superior

cs.CL↗

arXiv2025

Context-Dependent Autonomic Responses in Social Anxiety During Cognitive-Emotional Stress

Social anxiety disorder (SAD) is associated with heightened physiological arousal during socially evaluative situations, yet it remains unclear whether similar autonomic responses emerge during non-evaluative cognitive-emotional stress. This study investigated wearable electrodermal activity (EDA) responses in socially anxious (SA) and non-socially anxious (NSA) individuals during an emotionally salient 2-back working memory task involving facial expressions. Fifty participants (25 SA, 25 NSA) completed a resting-state baseline and task condition while EDA signals were acquired using a Shimmer3 GSR+ sensor. EDA features spanning tonic, phasic, sympathetic, spectral, and nonlinear domains were analyzed using mixed ANOVAs and complementary machine learning models. Results showed significant increases in autonomic arousal during task engagement across all participants, confirming that the task induced substantial sympathetic activation. However, no consistent between-group differences were observed, with only transient interaction effects emerging during the initial task phase. Machine learning analysis demonstrated above-chance discrimination between SA and NSA individuals using resting-state EDA (average AUC~=~0.73), whereas classification performance during task engagement declined to near-chance levels (average AUC~$\leq$~0.57). These findings suggest that cognitively demanding emotional tasks, in the absence of explicit social-evaluative threat, elicit comparable autonomic responses regardless of social anxiety status and may obscure subtle resting-state physiological differences between groups. More broadly, our findings highlight the context-dependent nature of wearable autonomic biomarkers for anxiety assessment and digital mental health monitoring. With this manuscript, we release both the code and data publicly.

q-bio.NC↗

arXiv2025

Geometric Preference Elicitation for Minimax Regret Optimization in Uncertainty Matroids

This paper presents an efficient preference elicitation framework for uncertain matroid optimization, where precise weight information is unavailable, but insights into possible weight values are accessible. The core innovation of our approach lies in its ability to systematically elicit user preferences, aligning the optimization process more closely with decision-makers' objectives. By incrementally querying preferences between pairs of elements, we iteratively refine the parametric uncertainty regions, leveraging the structural properties of matroids. Our method aims to achieve the exact optimum by reducing regret with a few elicitation rounds. Additionally, our approach avoids the computation of Minimax Regret and the use of Linear programming solvers at every iteration, unlike previous methods. Experimental results on four standard matroids demonstrate that our method reaches optimality more quickly and with fewer preference queries than existing techniques.

cs.LG↗

arXiv2024

Multimodal Fusion Learning with Dual Attention for Medical Imaging

Multimodal fusion learning has shown significant promise in classifying various diseases such as skin cancer and brain tumors. However, existing methods face three key limitations. First, they often lack generalizability to other diagnosis tasks due to their focus on a particular disease. Second, they do not fully leverage multiple health records from diverse modalities to learn robust complementary information. And finally, they typically rely on a single attention mechanism, missing the benefits of multiple attention strategies within and across various modalities. To address these issues, this paper proposes a dual robust information fusion attention mechanism (DRIFA) that leverages two attention modules, i.e. multi-branch fusion attention module and the multimodal information fusion attention module. DRIFA can be integrated with any deep neural network, forming a multimodal fusion learning framework denoted as DRIFA-Net. We show that the multi-branch fusion attention of DRIFA learns enhanced representations for each modality, such as dermoscopy, pap smear, MRI, and CT-scan, whereas multimodal information fusion attention module learns more refined multimodal shared representations, improving the network's generalization across multiple tasks and enhancing overall performance. Additionally, to estimate the uncertainty of DRIFA-Net predictions, we have employed an ensemble Monte Carlo dropout strategy. Extensive experiments on five publicly available datasets with diverse modalities demonstrate that our approach consistently outperforms state-of-the-art methods. The code is available at https://github.com/misti1203/DRIFA-Net.

cs.CV↗

arXiv2024

Study of $\beta^+$/EC-decay properties of $sd$ shell nuclei using nuclear shell model

Our study employs the nuclear shell model to systematically compute the half-lives of $\beta$ -decay for nuclei in the mass range of $A = 18-39$, encompassing the majority of $sd$ shell nuclei. This analysis utilizes the USDB and SDNN Hamiltonians. The theoretical outcomes contain calculations of various parameters such as $Q$ -values, half-lives, excitation energy, log$ft$ values, and branching ratios. We explore these results with axial-vector coupling constant for weak interactions, denoted as $g_A$$(= 1.27)$, and $\kappa$ value $(= 6289)$. We perform calculations of Gamow Teller matrix elements for 116 decay processes to calculate the quenching factor; we found a quenching factor of $q = 0.794\pm0.05 $ for the USDB interaction and $q = 0.815\pm0.04 $ for the SDNN interaction. We have also calculated superallowed transitions $0^+ \rightarrow 0^+$ for seven nuclei. Further, we have also included the electron capture phase space factor for the required nuclei to calculate the half-lives. This inclusion leads to small contribution in results, particularly for nuclei where electron capture (EC) plays a significant role. The overall results are in agreement with the experimental data.

nucl-th↗

arXiv2024

Citation-Based Summarization of Landmark Judgments

Landmark judgments are of prime importance in the Common Law System because of their exceptional jurisprudence and frequent references in other judgments. In this work, we leverage contextual references available in citing judgments to create an extractive summary of the target judgment. We evaluate the proposed algorithm on two datasets curated from the judgments of Indian Courts and find the results promising.

cs.CL↗

arXiv2024

Routing Algorithms

Routing algorithms play a crucial role in the efficient transmission of data within computer networks by determining the optimal paths for packet forwarding. This paper presents a comprehensive exploration of routing algorithms, focusing on their fundamental principles, classification, challenges, recent advancements, and practical applications. Beginning with an overview of the significance of routing in modern communication networks, the paper delves into the historical evolution of routing algorithms, tracing their development from early approaches to contemporary techniques. Key categories of routing algorithms, including distance vector, link-state, and path vector algorithms, are examined in detail, along with hybrid approaches that integrate multiple routing paradigms. Common challenges faced by routing algorithms, such as routing loops and scalability issues, are identified, and current research efforts aimed at addressing these challenges are discussed.

cs.NI↗

arXiv2024

Human Centered AI for Indian Legal Text Analytics

Legal research is a crucial task in the practice of law. It requires intense human effort and intellectual prudence to research a legal case and prepare arguments. Recent boom in generative AI has not translated to proportionate rise in impactful legal applications, because of low trustworthiness and and the scarcity of specialized datasets for training Large Language Models (LLMs). This position paper explores the potential of LLMs within Legal Text Analytics (LTA), highlighting specific areas where the integration of human expertise can significantly enhance their performance to match that of experts. We introduce a novel dataset and describe a human centered, compound AI system that principally incorporates human inputs for performing LTA tasks with LLMs.

cs.HC↗

arXiv2024

Infusing Knowledge into Large Language Models with Contextual Prompts

Knowledge infusion is a promising method for enhancing Large Language Models for domain-specific NLP tasks rather than pre-training models over large data from scratch. These augmented LLMs typically depend on additional pre-training or knowledge prompts from an existing knowledge graph, which is impractical in many applications. In contrast, knowledge infusion directly from relevant documents is more generalisable and alleviates the need for structured knowledge graphs while also being useful for entities that are usually not found in any knowledge graph. With this motivation, we propose a simple yet generalisable approach for knowledge infusion by generating prompts from the context in the input text. Our experiments show the effectiveness of our approach which we evaluate by probing the fine-tuned LLMs.

cs.CL↗

arXiv2023

Deep dive into language traits of AI-generated Abstracts

Generative language models, such as ChatGPT, have garnered attention for their ability to generate human-like writing in various fields, including academic research. The rapid proliferation of generated texts has bolstered the need for automatic identification to uphold transparency and trust in the information. However, these generated texts closely resemble human writing and often have subtle differences in the grammatical structure, tones, and patterns, which makes systematic scrutinization challenging. In this work, we attempt to detect the Abstracts generated by ChatGPT, which are much shorter in length and bounded. We extract the texts semantic and lexical properties and observe that traditional machine learning models can confidently detect these Abstracts.

cs.CL↗

arXiv2023

Shell-model study of $\beta^+$/EC-decay half-lives for $Z = 21-30$ nuclei

In the present work, we have reported $\beta^+$/EC-decay half-lives for $Z = 21 -30$ nuclei using large-scale shell-model. %and N $\leq$ A recent study shows that some proton-rich nuclei in this region belong to the island of inversion. We have performed calculations for these nuclei using KB3G effective interaction, while for Ni, Cu, and Zn nuclei we have used JUN45 effective interaction in the $f_{5/2}pg_{9/2}$ model space. The calculated quenching factors for $fp$ and $f_{5/2}pg_{9/2}$ space using KB3G and JUN45 effective interactions are also reported. Shell-model results of $\beta$-decay half-lives, excitation energies, log$ft$ values, and branching fractions are discussed and compared with the available experimental data. We have obtained a reasonable agreement with the available data.

nucl-th↗

arXiv2023

Lattice attack on group ring NTRU: The case of the dihedral group

Group ring NTRU (GR-NTRU) provides a general structure to design different variants of NTRU-like schemes by employing different groups. Although, most of the schemes in literature are built over cyclic groups, nonabelian groups can also be used. Coppersmith and Shamir in 1997 have suggested that noncommutativity may result in better security against some lattice attacks for some groups. Lattice attacks on the public key of NTRU-like cryptosystems try to retrieve the private key by solving the shortest vector problem (SVP) or its approximation in a lattice of a certain dimension, assuming the knowledge of the public key only. This paper shows that dihedral groups do not guarantee better security against this class of attacks. We prove that retrieving the private key is possible by solving the SVP in two lattices with half the dimension of the original lattice generated for GR-NTRU based on dihedral groups. The possibility of such an attack was mentioned by Yasuda et al.(IACR/2015/1170). In contrast to their proposed approach, we explicitly provide the lattice reduction without any structure theorem from the representation theory for finite groups. Furthermore, we demonstrate the effectiveness of our technique with experimental results.

cs.CR↗