arXiv subjects

Sarmistha Das

Publications and source records attributed to Sarmistha Das.

16 recordsLinked to original sources

arXiv2026

When Meaning Travels: A Granular Lens on Hybrid-MoE's Role in Idiomatic Understanding for Language Models

In the contemporary epoch of multilingual education, learning idioms provides a fascinating gateway towards creativity, cultural values, historical context, and diverse perspectives inherent to various linguistic traditions. This paper showcases the navigation of retaining figurative and cultural semantics in low-resource Southeast Asian languages such as Hindi, Bengali, and Thai, where culturally rich idioms pose significant obstacles for computational modeling and cross-linguistic transfer due to their deep metaphorical complexity. To tackle such complexity, we present Varnika, a reconstructed multimodal idiom corpus comprising 3,533 multilingual idioms, enriched with seven idiomatic tones aligned with both textual and visual representations. Additionally, to infer informative idiomatic understanding, we introduce a Hybrid Mixture-of-Experts (HybridMoE) framework that embeds multiple idiomatic expert opinions while mitigating expert sparsity by integrating outputs from both selected and unselected experts through controlled hybridization, further augmented with Idiomatic Property Signals via masked multimodal embeddings. To analyze the performance across multiple dimensions, we propose the IDIO-TONE and Idiomatic Validation Score, a three-stage evaluation pipeline measuring (i) literal translation fidelity, (ii) visual-semantic alignment, and (iii) idiomatic meaning retention. Empirical evaluations highlight that HybridMoE achieves 5--6\% performance gains across advanced vision language models, demonstrating improved representation of figurative language and culturally embedded meaning in multilingual multimodal settings

cs.CL↗

arXiv2026

Low-Temperature Suppression of Intertwined Orders in La$_{1/3}$Sr$_{2/3}$FeO$_{3}$ Thin Films

The strong coupling between spin, charge, and lattice degrees of freedom in perovskite oxides leads to an array of exotic phenomena, giving these materials rich phase diagrams that can include coupled orders. This is exemplified by the A-site doped ferrite La$_{1/3}$Sr$_{2/3}$FeO$_{3}$ (LSFO), which exhibits a coupled paramagnetic-antiferromagnetic and charge ordering phase transition at $\sim$190 K that has been well studied in thin films, bulk, and polycrystalline samples. However, the low temperature behavior of LSFO thin films below $\sim$100 K has not been thoroughly explored. This work uses several X-ray scattering and spectroscopy techniques to directly probe LSFO's magnetic and charge order down to low temperature. Using resonant X-ray scattering, we observe a complete suppression of LSFO's known antiferromagnetic and charge order below $\sim$25 K. Further spectroscopy and coherent scattering measurements provide insight into LSFO's electronic structure and domain dynamics in this new low temperature phase, and we propose possible explanations for the observed order suppression based on reduced dimensionality of domains in our thin films. Our findings provide insight into the effects of competing interactions in strongly correlated materials, particularly those with coupled orders.

cond-mat.str-el↗

arXiv2026

FIND: Toward Multimodal Financial Reasoning and Question Answering for Indic Languages

Financial decision-making in multilingual settings demands accurate numerical reasoning grounded in diverse modalities, yet existing benchmarks largely overlook this high-stakes, real-world challenge, especially for Indic languages. We introduce FinVQA, a benchmark for evaluating financial numerical and multimodal reasoning in multilingual Indic contexts. FinVQA spans English, Hindi, Bengali, Marathi, Gujarati, and Tamil, and comprises 18,900 samples across 14 financial domains. The dataset captures diverse reasoning paradigms under realistic constraints, and is structured across three difficulty levels (easy, moderate, hard) and four question formats: multiple choice, fill-in-the-blank, table matching, and true/false. To address these challenges, we propose FIND, a framework that combines supervised fine-tuning with constraint-aware decoding to promote faithful numerical reasoning, robust multimodal grounding, and structured decision-making. Together, FinVQA and FIND establish a rigorous evaluation and modeling paradigm for high-stakes multilingual multimodal financial reasoning.

cs.CL↗

arXiv2026

When Meaning Isn't Literal: Exploring Idiomatic Meaning Across Languages and Modalities

Idiomatic reasoning, deeply intertwined with metaphor and culture, remains a blind spot for contemporary language models, whose progress skews toward surface-level lexical and semantic cues. For instance, the Bengali idiom \textit{\foreignlanguage{bengali}{\char"0986\char"0999\char"09CD\char"0997\char"09C1 \char"09B0 \char"09AB\char"09B2 \char"099F\char"0995}} (angur fol tok, ``grapes are sour''): it encodes denial-driven rationalization, yet naive models latch onto the literal fox-and-grape imagery. Addressing this oversight, we present ``Mediom,'' a multilingual, multimodal idiom corpus of 3,533 Hindi, Bengali, and Thai idioms, each paired with gold-standard explanations, cross-lingual translations, and carefully aligned text--image representations. We benchmark both large language models (textual reasoning) and vision-language models (figurative disambiguation) on Mediom, exposing systematic failures in metaphor comprehension. To mitigate these gaps, we propose ``HIDE,'' a Hinting-based Idiom Explanation framework that leverages error-feedback retrieval and targeted diagnostic cues for iterative reasoning refinement. Collectively, Mediom and HIDE establish a rigorous test bed and methodology for culturally grounded, multimodal idiom understanding embedded with reasoning hints in next-generation AI systems.

cs.CL↗

arXiv2025

Talk, Snap, Complain: Validation-Aware Multimodal Expert Framework for Fine-Grained Customer Grievances

Existing approaches to complaint analysis largely rely on unimodal, short-form content such as tweets or product reviews. This work advances the field by leveraging multimodal, multi-turn customer support dialogues, where users often share both textual complaints and visual evidence (e.g., screenshots, product photos) to enable fine-grained classification of complaint aspects and severity. We introduce VALOR, a Validation-Aware Learner with Expert Routing, tailored for this multimodal setting. It employs a multi-expert reasoning setup using large-scale generative models with Chain-of-Thought (CoT) prompting for nuanced decision-making. To ensure coherence between modalities, a semantic alignment score is computed and integrated into the final classification through a meta-fusion strategy. In alignment with the United Nations Sustainable Development Goals (UN SDGs), the proposed framework supports SDG 9 (Industry, Innovation and Infrastructure) by advancing AI-driven tools for robust, scalable, and context-aware service infrastructure. Further, by enabling structured analysis of complaint narratives and visual context, it contributes to SDG 12 (Responsible Consumption and Production) by promoting more responsive product design and improved accountability in consumer services. We evaluate VALOR on a curated multimodal complaint dataset annotated with fine-grained aspect and severity labels, showing that it consistently outperforms baseline models, especially in complex complaint scenarios where information is distributed across text and images. This study underscores the value of multimodal interaction and expert validation in practical complaint understanding systems. Resources related to data and codes are available here: https://github.com/sarmistha-D/VALOR

cs.CL↗

arXiv2025

Fin-Ally: Pioneering the Development of an Advanced, Commonsense-Embedded Conversational AI for Money Matters

The exponential technological breakthrough of the FinTech industry has significantly enhanced user engagement through sophisticated advisory chatbots. However, large-scale fine-tuning of LLMs can occasionally yield unprofessional or flippant remarks, such as ``With that money, you're going to change the world,'' which, though factually correct, can be contextually inappropriate and erode user trust. The scarcity of domain-specific datasets has led previous studies to focus on isolated components, such as reasoning-aware frameworks or the enhancement of human-like response generation. To address this research gap, we present Fin-Solution 2.O, an advanced solution that 1) introduces the multi-turn financial conversational dataset, Fin-Vault, and 2) incorporates a unified model, Fin-Ally, which integrates commonsense reasoning, politeness, and human-like conversational dynamics. Fin-Ally is powered by COMET-BART-embedded commonsense context and optimized with a Direct Preference Optimization (DPO) mechanism to generate human-aligned responses. The novel Fin-Vault dataset, consisting of 1,417 annotated multi-turn dialogues, enables Fin-Ally to extend beyond basic account management to provide personalized budgeting, real-time expense tracking, and automated financial planning. Our comprehensive results demonstrate that incorporating commonsense context enables language models to generate more refined, textually precise, and professionally grounded financial guidance, positioning this approach as a next-generation AI solution for the FinTech sector. Dataset and codes are available at: https://github.com/sarmistha-D/Fin-Ally

cs.AI↗

arXiv2025

Unlocking Financial Insights: An advanced Multimodal Summarization with Multimodal Output Framework for Financial Advisory Videos

The dynamic propagation of social media has broadened the reach of financial advisory content through podcast videos, yet extracting insights from lengthy, multimodal segments (30-40 minutes) remains challenging. We introduce FASTER (Financial Advisory Summariser with Textual Embedded Relevant images), a modular framework that tackles three key challenges: (1) extracting modality-specific features, (2) producing optimized, concise summaries, and (3) aligning visual keyframes with associated textual points. FASTER employs BLIP for semantic visual descriptions, OCR for textual patterns, and Whisper-based transcription with Speaker diarization as BOS features. A modified Direct Preference Optimization (DPO)-based loss function, equipped with BOS-specific fact-checking, ensures precision, relevance, and factual consistency against the human-aligned summary. A ranker-based retrieval mechanism further aligns keyframes with summarized content, enhancing interpretability and cross-modal coherence. To acknowledge data resource scarcity, we introduce Fin-APT, a dataset comprising 470 publicly accessible financial advisory pep-talk videos for robust multimodal research. Comprehensive cross-domain experiments confirm FASTER's strong performance, robustness, and generalizability when compared to Large Language Models (LLMs) and Vision-Language Models (VLMs). By establishing a new standard for multimodal summarization, FASTER makes financial advisory content more accessible and actionable, thereby opening new avenues for research. The dataset and code are available at: https://github.com/sarmistha-D/FASTER

cs.CV↗

arXiv2025

When Words Can't Capture It All: Towards Video-Based User Complaint Text Generation with Multimodal Video Complaint Dataset

While there exists a lot of work on explainable complaint mining, articulating user concerns through text or video remains a significant challenge, often leaving issues unresolved. Users frequently struggle to express their complaints clearly in text but can easily upload videos depicting product defects (e.g., vague text such as `worst product' paired with a 5-second video depicting a broken headphone with the right earcup). This paper formulates a new task in the field of complaint mining to aid the common users' need to write an expressive complaint, which is Complaint Description from Videos (CoD-V) (e.g., to help the above user articulate her complaint about the defective right earcup). To this end, we introduce ComVID, a video complaint dataset containing 1,175 complaint videos and the corresponding descriptions, also annotated with the emotional state of the complainer. Additionally, we present a new complaint retention (CR) evaluation metric that discriminates the proposed (CoD-V) task against standard video summary generation and description tasks. To strengthen this initiative, we introduce a multimodal Retrieval-Augmented Generation (RAG) embedded VideoLLaMA2-7b model, designed to generate complaints while accounting for the user's emotional state. We conduct a comprehensive evaluation of several Video Language Models on several tasks (pre-trained and fine-tuned versions) with a range of established evaluation metrics, including METEOR, perplexity, and the Coleman-Liau readability score, among others. Our study lays the foundation for a new research direction to provide a platform for users to express complaints through video. Dataset and resources are available at: https://github.com/sarmistha-D/CoD-V.

cs.CV↗

arXiv2025

Enhanced spin-to-charge conversion in La$_{0.67}$Sr$_{0.33}$MnO$_3$/NdNiO$_3$ bilayers at the nickelate metal-insulator phase transition

Phase transition materials such as NdNiO3 (NNO) when coupled with low damping ferromagnets such as La$_{0.67}$Sr$_{0.33}$MnO$_3$ (LSMO) can lead to new multi-functional material systems harnessing the interplay of charge, spin and orbital degrees of freedom. In this study, we probe the evolution of the spin-to-charge conversion in epitaxial all-oxide LSMO (12 nm)/NNO (4, 8, and 16 nm) bilayers. Using spin pumping ferromagnetic resonance we track the spin-charge conversion in the NNO layer through the paramagnetic metal to antiferromagnetic insulator transition and observe a pronounced enhancement of the inverse spin Hall effect signal at the onset of this transition. We attribute this enhancement to the electronic and magnetic disorder in NNO at the first-order phase transition, thereby providing insights into the mechanism of spin transport through the phase transition. The tunability of spin charge conversion in this low damping bilayer system offers a pathway for developing multifunctional, energy-efficient spintronic devices.

cond-mat.mtrl-sci↗

arXiv2025

Deciphering the complaint aspects: Towards an aspect-based complaint identification model with video complaint dataset in finance

In today's competitive marketing landscape, effective complaint management is crucial for customer service and business success. Video complaints, integrating text and image content, offer invaluable insights by addressing customer grievances and delineating product benefits and drawbacks. However, comprehending nuanced complaint aspects within vast daily multimodal financial data remains a formidable challenge. Addressing this gap, we have curated a proprietary multimodal video complaint dataset comprising 433 publicly accessible instances. Each instance is meticulously annotated at the utterance level, encompassing five distinct categories of financial aspects and their associated complaint labels. To support this endeavour, we introduce Solution 3.0, a model designed for multimodal aspect-based complaint identification task. Solution 3.0 is tailored to perform three key tasks: 1) handling multimodal features ( audio and video), 2) facilitating multilabel aspect classification, and 3) conducting multitasking for aspect classifications and complaint identification parallelly. Solution 3.0 utilizes a CLIP-based dual frozen encoder with an integrated image segment encoder for global feature fusion, enhanced by contextual attention (ISEC) to improve accuracy and efficiency. Our proposed framework surpasses current multimodal baselines, exhibiting superior performance across nearly all metrics by opening new ways to strengthen appropriate customer care initiatives and effectively assisting individuals in resolving their problems.

cs.CV↗

arXiv2025

Weyl and Dirac Semimetals for Thermoelectric Applications

Weyl and Dirac semimetals, characterized by their unique band structures with linear energy dispersion (E vs k) near the Fermi level (EF), have emerged as promising candidates for next-generation technology based on thermoelectric materials. Their exceptional electronic properties, notably high carrier mobility and substantial Berry curvature, offer the potential to surmount the limitations inherent in conventional thermoelectric materials. A comprehensive understanding of the fundamental physics underlying these materials is essential. This chapter mainly focused into the topological properties and distinctive electronic band structures of Weyl and Dirac semimetals, providing a theoretical framework for comprehending their thermoelectric transport properties such as Seebeck coefficients, electrical and thermal conductivity. The pivotal role of Berry curvature in enhancing Seebeck coefficients while reducing thermal conductivity is a key focus. Experimental advancements in synthesizing single crystals and characterizing these materials have been significant. Recent development in material growth and characterization techniques have propelled research forward. The intricate relationship between material properties, such as carrier concentration, electronic bandgap, and crystal structure, and thermoelectric performance is explored. Realizing the potential of Weyl and Dirac semimetals for practical thermoelectric applications necessitates overcoming specific challenges. This chapter outlines strategies to optimize thermoelectric figures of merit (ZT) through band engineering, carrier doping, and nanostructuring. Moreover, the exploration of hybrid materials and heterostructures offers promising avenues for enhancing thermoelectric performance for renewable energy applications.

cond-mat.mtrl-sci↗

arXiv2024

Temperature dependent spin dynamics in La$_{0.67}$Sr$_{0.33}$MnO$_3$/Pt bilayers

Complex ferromagnetic oxides such as La$_{0.67}$Sr$_{0.33}$MnO$_3$ (LSMO) offer pathways for creating energy efficient spintronic devices with new functionalities. LSMO exhibits high-temperature ferromagnetism, half metallicity, sharp resonance linewidth, low damping and a large anisotropic magnetoresistance response. Combined with Pt, a proven material with high spin-charge conversion efficiency, LSMO can be used to create robust nano-oscillators for neuromorphic computing. Ferromagnetic resonance (FMR) and device level spin-pumping FMR measurements are performed to investigate the magnetization dynamics and spin transport in NdGaO3(110)/LSMO(15 nm)/Pt(0 and 5 nm) thin films ranging from 300K to 90K and compare the device performance with Py(7 nm)/Pt(5 nm) sample. The spin current pumped into Pt is quantified to determine the temperature dependent influence of interfacial interactions. The generated spin current in the micro-device is maximum at 170K for the optimally grown LSMO/Pt films. Additionally, this bilayer system exhibits low magnetic Gilbert damping (0.002), small linewidth (12 Oe) and a large spin Hall angle ($\approx$ 3.2%) at 170K. By fine-tuning the LSMO/Pt interface quality and integrating it into the device structure, the system exhibits a fourfold enhancement in signal output for LSMO/Pt devices compared to the Pt/Py system. Such robust device level performance can pave way for energy-efficient spintronic based devices.

cond-mat.mtrl-sci↗

arXiv2023

Disentangling transport mechanisms in a correlated oxide by photoinduced charge injection

We present a novel heterostructured approach to disentangle the mechanism of electrical transport of the strongly correlated PrNiO3, by placing the nickelate under the photoconductor CdS. This enables the injection of carriers into PrNiO3 in a controlled way, which can be used to interrogate its intrinsic transport mechanism. We find a non-volatile resistance decrease when illuminating the system at temperatures below the PrNiO3 metal-insulator transition. The photoinduced change becomes more volatile as the temperature increases. These data help understand the intrinsic transport properties of the nickelate-CdS bilayer. Together with data from a bare PrNiO3 film, we find that the transport mechanism includes a combination of mechanisms including both thermal activation and variable range hopping. At low temperatures without photoinduced carriers the transport is governed by hopping, while at higher temperatures and intense illumination the activation mechanism becomes relevant. This work shows a new way to optically control the low-temperature resistance of PrNiO3.

cond-mat.str-el↗

arXiv2021

Querying multiple sets of $p$-values through composed hypothesis testing

Motivation: Combining the results of different experiments to exhibit complex patterns or to improve statistical power is a typical aim of data integration. The starting point of the statistical analysis often comes as sets of p-values resulting from previous analyses, that need to be combined in a flexible way to explore complex hypotheses, while guaranteeing a low proportion of false discoveries. Results: We introduce the generic concept of composed hypothesis, which corresponds to an arbitrary complex combination of simple hypotheses. We rephrase the problem of testing a composed hypothesis as a classification task, and show that finding items for which the composed null hypothesis is rejected boils down to fitting a mixture model and classify the items according to their posterior probabilities. We show that inference can be efficiently performed and provide a thorough classification rule to control for type I error. The performance and the usefulness of the approach are illustrated on simulations and on two different applications. The method is scalable, does not require any parameter tuning, and provided valuable biological insight on the considered application cases. Availability: The QCH methodology is implemented in the qch R package hosted on CRAN.

stat.ME↗

arXiv2021

Probing the evolution of electronic phase-coexistence in complex systems by terahertz radiation

In complex oxides, the electrons under the influence of competing energetics are the cornerstone of coexistence (or phase-separation) of two or more electronic/magnetic phases in same structural configuration. Probing of growth and evolution of such phase-coexistence state is crucial to determine the correct mechanism of related phase-transition. Here, we demonstrate the combination of terahertz (THz) time-domain spectroscopy and DC transport as a novel strategy to probe the electronic phase-coexistence. This is demonstrated in disorder controlled phase-separated rare-earth nickelate thin films which exhibit metal-insulator transition in dc conductivity at around 180 K but lack this transition in terahertz (THz) dynamics conductivity down to low temperature. Such pronounced disparity exploits two extreme attributes: i) enormous sensitivity of THz radiation to a spatial range of its wavelength-compatible electronic inhomogeneities and ii) insensitivity to a range beyond the size of its wavelength. This feature is generic in nature (sans a photo-induced effect), depends solely on the size of insulating/metallic clusters and formulates a methodology with unique sensitivity to investigate electronic phase-coexistence and phase transition of any material system.

cond-mat.str-el↗

arXiv2020

Critical community size for COVID-19 -- a model based approach to provide a rationale behind the lockdown

Background: Restrictive mass quarantine or lockdown has been implemented as the most important controlling measure to fight against COVID-19. Many countries have enforced 2 - 4 weeks' lockdown and are extending the period depending on their current disease scenario. Most probably the 14-day period of estimated communicability of COVID-19 prompted such decision. But the idea that, if the susceptible population drops below certain threshold, the infection would naturally die out in small communities after a fixed time (following the outbreak), unless the disease is reintroduced from outside, was proposed by Bartlett in 1957. This threshold was termed as Critical Community Size (CCS). Methods: We propose an SEIR model that explains COVID-19 disease dynamics. Using our model, we have calculated country-specific expected time to extinction (TTE) and CCS that would essentially determine the ideal number of lockdown days required and size of quarantined population. Findings: With the given country-wise rates of death, recovery and other parameters, we have identified that, if at a place the total number of susceptible population drops below CCS, infection will cease to exist after a period of TTE days, unless it is introduced from outside. But the disease will almost die out much sooner. We have calculated the country-specific estimate of the ideal number of lockdown days. Thus, smaller lockdown phase is sufficient to contain COVID-19. On a cautionary note, our model indicates another rise in infection almost a year later but on a lesser magnitude.

q-bio.PE↗