<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="khanal-2026-zeror">
    <titleInfo>
        <title>ZeroR@CHiPSAL 2026: Two-Stage Vision-Language Adaptation with Contrastive Learning for Nepali Meme Classification</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Nitiz</namePart>
        <namePart type="family">Khanal</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <originInfo>
        <dateIssued>2026-05</dateIssued>
    </originInfo>
    <typeOfResource>text</typeOfResource>
    <relatedItem type="host">
        <titleInfo>
            <title>Proceedings of the Second workshop on Challenges in Processing South Asian Languages (CHiPSAL2026)</title>
        </titleInfo>
        <name type="personal">
            <namePart type="given">Kengatharaiyer</namePart>
            <namePart type="family">Sarveswaran</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <name type="personal">
            <namePart type="given">Ashwini</namePart>
            <namePart type="family">Vaidya</namePart>
            <role>
                <roleTerm authority="marcrelator" type="text">editor</roleTerm>
            </role>
        </name>
        <originInfo>
            <publisher>ELRA Language Resources Association (ELRA)</publisher>
            <place>
                <placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
            </place>
        </originInfo>
        <genre authority="marcgt">conference publication</genre>
    </relatedItem>
    <abstract>This paper presents our system for the CHiPSAL 2026 shared task on multimodal hate speech and sentiment detection in Nepali memes. We address both subtasks: binary hate speech classification and three-class sentiment analysis. Our approach adapts the Robust Adaptation of Hateful Meme Detection (RA-HMD) framework using Qwen3-VL-8B-Instruct, a state-of-the-art vision-language model with native Devanagari support. We employ a two-stage training pipeline: (1) LoRA fine-tuning with an MLP projection head for generative classification, and (2) contrastive backbone fine-tuning with supervised InfoNCE loss. We handle class imbalance through minority oversampling, image augmentation, and focal loss. At inference, we ensemble Stage 1 token probabilities with Stage 2 classifier scores using validation-tuned weights. Our end-to-end approach eliminates error propagation from separate OCR and translation pipelines by leveraging the model’s native Devanagari understanding. Our system achieved 2nd place on hate speech detection (F1: 0.797) and 4th place on sentiment analysis (F1: 0.518). We provide detailed ablations, error analysis, and insights into adapting large vision-language models for low-resource South Asian languages.</abstract>
    <identifier type="citekey">khanal-2026-zeror</identifier>
    <identifier type="doi">10.63317/5nypaugdk6kz</identifier>
    <location>
        <url>https://aclanthology.org/2026.chipsal-1.28/</url>
    </location>
    <part>
        <date>2026-05</date>
        <extent unit="page">
            <start>275</start>
            <end>283</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T ZeroR@CHiPSAL 2026: Two-Stage Vision-Language Adaptation with Contrastive Learning for Nepali Meme Classification
%A Khanal, Nitiz
%Y Sarveswaran, Kengatharaiyer
%Y Vaidya, Ashwini
%S Proceedings of the Second workshop on Challenges in Processing South Asian Languages (CHiPSAL2026)
%D 2026
%8 May
%I ELRA Language Resources Association (ELRA)
%C Palma de Mallorca, Spain
%F khanal-2026-zeror
%X This paper presents our system for the CHiPSAL 2026 shared task on multimodal hate speech and sentiment detection in Nepali memes. We address both subtasks: binary hate speech classification and three-class sentiment analysis. Our approach adapts the Robust Adaptation of Hateful Meme Detection (RA-HMD) framework using Qwen3-VL-8B-Instruct, a state-of-the-art vision-language model with native Devanagari support. We employ a two-stage training pipeline: (1) LoRA fine-tuning with an MLP projection head for generative classification, and (2) contrastive backbone fine-tuning with supervised InfoNCE loss. We handle class imbalance through minority oversampling, image augmentation, and focal loss. At inference, we ensemble Stage 1 token probabilities with Stage 2 classifier scores using validation-tuned weights. Our end-to-end approach eliminates error propagation from separate OCR and translation pipelines by leveraging the model’s native Devanagari understanding. Our system achieved 2nd place on hate speech detection (F1: 0.797) and 4th place on sentiment analysis (F1: 0.518). We provide detailed ablations, error analysis, and insights into adapting large vision-language models for low-resource South Asian languages.
%R 10.63317/5nypaugdk6kz
%U https://aclanthology.org/2026.chipsal-1.28/
%U https://doi.org/10.63317/5nypaugdk6kz
%P 275-283

Markdown (Informal)

[ZeroR@CHiPSAL 2026: Two-Stage Vision-Language Adaptation with Contrastive Learning for Nepali Meme Classification](https://aclanthology.org/2026.chipsal-1.28/) (Khanal, CHiPSAL 2026)

• ZeroR@CHiPSAL 2026: Two-Stage Vision-Language Adaptation with Contrastive Learning for Nepali Meme Classification (Khanal, CHiPSAL 2026)

ACL

• Nitiz Khanal. 2026. ZeroR@CHiPSAL 2026: Two-Stage Vision-Language Adaptation with Contrastive Learning for Nepali Meme Classification. In Proceedings of the Second workshop on Challenges in Processing South Asian Languages (CHiPSAL2026), pages 275–283, Palma de Mallorca, Spain. ELRA Language Resources Association (ELRA).