<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Publications on Arya Suneesh</title><link>https://aryasuneesh.github.io/publications/</link><description>Recent content in Publications on Arya Suneesh</description><generator>Hugo -- 0.147.2</generator><language>en</language><lastBuildDate>Sun, 12 Oct 2025 00:00:00 +0000</lastBuildDate><atom:link href="https://aryasuneesh.github.io/publications/index.xml" rel="self" type="application/rss+xml"/><item><title>QCNN-MFND: A Novel Quantum CNN Framework for Multimodal Fake News Detection in Social Media</title><link>https://aryasuneesh.github.io/publications/paper3/</link><pubDate>Sun, 12 Oct 2025 00:00:00 +0000</pubDate><guid>https://aryasuneesh.github.io/publications/paper3/</guid><description>A hybrid quantum-classical framework that fuses text and image features through quantum convolutional neural networks, achieving 88.52% accuracy on GossipCop with 65% fewer false negatives than classical baselines.</description></item><item><title>TIFIN at CheckThat! 2025: Cross-Lingual Subjectivity Classification in News through Monolingual, Multilingual, and Zero-Shot Learning</title><link>https://aryasuneesh.github.io/publications/paper4/</link><pubDate>Fri, 12 Sep 2025 00:00:00 +0000</pubDate><guid>https://aryasuneesh.github.io/publications/paper4/</guid><description>A comprehensive approach to cross-lingual subjectivity classification using transformer-based models with resampling and class-weighting techniques, achieving strong performance across monolingual, multilingual, and zero-shot settings in five languages.</description></item><item><title>TIFIN at CheckThat! 2025: Reasoning-Guided Claim Normalization for Noisy Multilingual Social Media Posts</title><link>https://aryasuneesh.github.io/publications/paper1/</link><pubDate>Fri, 12 Sep 2025 00:00:00 +0000</pubDate><guid>https://aryasuneesh.github.io/publications/paper1/</guid><description>A reasoning-guided approach to claim normalization using 5W1H decomposition and Qwen3-14B fine-tuning, achieving 41.3% relative improvement in METEOR scores over baseline configurations across 20 languages.</description></item><item><title>TIFIN India at SemEval-2025: Harnessing Translation to Overcome Multilingual IR Challenges in Fact-Checked Claim Retrieval</title><link>https://aryasuneesh.github.io/publications/paper2/</link><pubDate>Wed, 23 Apr 2025 00:00:00 +0000</pubDate><guid>https://aryasuneesh.github.io/publications/paper2/</guid><description>A novel approach to multilingual fact-checked claim retrieval using LLM-based translation and two-stage retrieval architecture, achieving high performance on both monolingual and crosslingual datasets.</description></item></channel></rss>