Selected work

Representative Papers

Multilingual Conversational AI for Financial Assistance: Bridging Language Barriers in Indian FinTech

Dec 01, 2025

Financial inclusion in India is hindered by extreme linguistic diversity—31 major languages coexist, yet only 10% of the population is proficient in English, creating significant barriers for low-resource language users in fintech interactions. Method: We propose a multilingual conversational AI system tailored for financial services, featuring a novel multi-agent architecture that jointly orchestrates language identification, functional routing, and response generation. The system natively supports code-mixed languages (e.g., Hinglish), integrates multilingual large language models, lightweight function routing, and domain-customized response generation, and is optimized on real-world financial dialogue data for semantic fidelity and inference efficiency. Contribution/Results: Deployment demonstrates substantial improvement in user engagement, with only a 4–8% increase in end-to-end latency. The system achieves high stability and computational efficiency across multilingual financial tasks, effectively overcoming longstanding resource scarcity constraints in low-resource language fintech applications.

0 citationsRead paper

Reasoning-Guided Claim Normalization for Noisy Multilingual Social Media Posts

Nov 07, 2025

This paper addresses the challenge of claim normalization in multilingual social media posts—transforming noisy, unstructured user-generated content into clear, verifiable statements to support multilingual misinformation detection—without requiring multilingual annotated data. Methodologically, it introduces a cross-lingual claim standardization framework based on a question-answering decomposition schema (Who/What/Where/When/Why/How), trained exclusively on English data. The approach integrates Qwen3-14B with LoRA fine-tuning and incorporates intra-post deduplication, token-level recall filtering, and retrieval-augmented few-shot inference. Experiments span 20 languages, achieving a peak METEOR score of 41.16—outperforming baseline methods by an average of 41.3%. Results demonstrate strong cross-lingual generalization and practical efficacy, significantly advancing state-of-the-art performance in zero-shot multilingual claim normalization.

0 citationsRead paper

TIFIN India at SemEval-2025: Harnessing Translation to Overcome Multilingual IR Challenges in Fact-Checked Claim Retrieval

Apr 23, 2025

To address the challenge of retrieving verified factual claims in both monolingual and cross-lingual settings for misinformation governance, this paper proposes a two-stage retrieval framework: initial retrieval via a fine-tuned semantic embedding model, followed by LLM-driven fine-grained re-ranking. We present the first empirical validation of LLM-based translation for cross-lingual fact-checking retrieval, and introduce LLM-assisted cross-lingual semantic alignment alongside lightweight inference optimization techniques—enabling efficient, reproducible deployment on consumer-grade GPUs. Experiments show that our method achieves Success@10 of 0.938 on monolingual test sets and 0.810 on cross-lingual ones, significantly outperforming existing baselines. The approach offers a practical, resource-efficient solution for multilingual fact-checking in low-resource scenarios.

0 citationsRead paper

Recent publications

Latest Papers

Multilingual Conversational AI for Financial Assistance: Bridging Language Barriers in Indian FinTech

Dec 01, 2025

Financial inclusion in India is hindered by extreme linguistic diversity—31 major languages coexist, yet only 10% of the population is proficient in English, creating significant barriers for low-resource language users in fintech interactions. Method: We propose a multilingual conversational AI system tailored for financial services, featuring a novel multi-agent architecture that jointly orchestrates language identification, functional routing, and response generation. The system natively supports code-mixed languages (e.g., Hinglish), integrates multilingual large language models, lightweight function routing, and domain-customized response generation, and is optimized on real-world financial dialogue data for semantic fidelity and inference efficiency. Contribution/Results: Deployment demonstrates substantial improvement in user engagement, with only a 4–8% increase in end-to-end latency. The system achieves high stability and computational efficiency across multilingual financial tasks, effectively overcoming longstanding resource scarcity constraints in low-resource language fintech applications.

0 citationsRead paper

Reasoning-Guided Claim Normalization for Noisy Multilingual Social Media Posts

Nov 07, 2025

This paper addresses the challenge of claim normalization in multilingual social media posts—transforming noisy, unstructured user-generated content into clear, verifiable statements to support multilingual misinformation detection—without requiring multilingual annotated data. Methodologically, it introduces a cross-lingual claim standardization framework based on a question-answering decomposition schema (Who/What/Where/When/Why/How), trained exclusively on English data. The approach integrates Qwen3-14B with LoRA fine-tuning and incorporates intra-post deduplication, token-level recall filtering, and retrieval-augmented few-shot inference. Experiments span 20 languages, achieving a peak METEOR score of 41.16—outperforming baseline methods by an average of 41.3%. Results demonstrate strong cross-lingual generalization and practical efficacy, significantly advancing state-of-the-art performance in zero-shot multilingual claim normalization.

0 citationsRead paper

TIFIN India at SemEval-2025: Harnessing Translation to Overcome Multilingual IR Challenges in Fact-Checked Claim Retrieval

Apr 23, 2025

To address the challenge of retrieving verified factual claims in both monolingual and cross-lingual settings for misinformation governance, this paper proposes a two-stage retrieval framework: initial retrieval via a fine-tuned semantic embedding model, followed by LLM-driven fine-grained re-ranking. We present the first empirical validation of LLM-based translation for cross-lingual fact-checking retrieval, and introduce LLM-assisted cross-lingual semantic alignment alongside lightweight inference optimization techniques—enabling efficient, reproducible deployment on consumer-grade GPUs. Experiments show that our method achieves Success@10 of 0.938 on monolingual test sets and 0.810 on cross-lingual ones, significantly outperforming existing baselines. The approach offers a practical, resource-efficient solution for multilingual fact-checking in low-resource scenarios.

0 citationsRead paper