FlaMe NLP Research Lab
@flamenlp.bsky.social
Fueling NLP with Passion! Research in #NLP led by Dr. Shad Akhtar @ IIIT Delhi, India
Our findings show a significant improvement in classifying mental health symptoms expressed in memes, highlighting the importance of integrating commonsense understanding in AI models.
Check out the full paper here: arxiv.org/abs/2501.15321
#TheWebConference #MentalHealth #NLProc #Research
3/3
Check out the full paper here: arxiv.org/abs/2501.15321
#TheWebConference #MentalHealth #NLProc #Research
3/3
Figurative-cum-Commonsense Knowledge Infusion for Multimodal Mental Health Meme Classification
The expression of mental health symptoms through non-traditional means, such as memes, has gained remarkable attention over the past few years, with users often highlighting their mental health strugg...
arxiv.org
January 29, 2025 at 4:46 AM
Our findings show a significant improvement in classifying mental health symptoms expressed in memes, highlighting the importance of integrating commonsense understanding in AI models.
Check out the full paper here: arxiv.org/abs/2501.15321
#TheWebConference #MentalHealth #NLProc #Research
3/3
Check out the full paper here: arxiv.org/abs/2501.15321
#TheWebConference #MentalHealth #NLProc #Research
3/3
AxiOM Dataset: A novel collection derived from the GAD questionnaire, categorizing memes into six fine-grained anxiety symptoms.
M3H Framework: A commonsense and domain-enriched approach to improve Multimodal Language Models' ability to interpret figurative language and commonsense knowledge.
2/3
M3H Framework: A commonsense and domain-enriched approach to improve Multimodal Language Models' ability to interpret figurative language and commonsense knowledge.
2/3
January 29, 2025 at 4:46 AM
AxiOM Dataset: A novel collection derived from the GAD questionnaire, categorizing memes into six fine-grained anxiety symptoms.
M3H Framework: A commonsense and domain-enriched approach to improve Multimodal Language Models' ability to interpret figurative language and commonsense knowledge.
2/3
M3H Framework: A commonsense and domain-enriched approach to improve Multimodal Language Models' ability to interpret figurative language and commonsense knowledge.
2/3
We’re looking forward to presenting this research at #NAACL2025 and sharing our contributions with the NLP community.
Stay tuned for the preprint, dataset, and code—coming soon on our lab’s website: FLaMe.nlp Research Lab.
#NAACL2025 #NLProc
2/2
Stay tuned for the preprint, dataset, and code—coming soon on our lab’s website: FLaMe.nlp Research Lab.
#NAACL2025 #NLProc
2/2
January 23, 2025 at 8:53 AM
We’re looking forward to presenting this research at #NAACL2025 and sharing our contributions with the NLP community.
Stay tuned for the preprint, dataset, and code—coming soon on our lab’s website: FLaMe.nlp Research Lab.
#NAACL2025 #NLProc
2/2
Stay tuned for the preprint, dataset, and code—coming soon on our lab’s website: FLaMe.nlp Research Lab.
#NAACL2025 #NLProc
2/2
2/n This work is with
Authors: Abdullah Mazhar, Zuhair Hasan Shaik, Aseem Srivastava, Polly Ruhnke, Lavanya Vaddavalli, Sri Keshav Katragadda, Shweta Yadav, Md Shad Akhtar
Authors: Abdullah Mazhar, Zuhair Hasan Shaik, Aseem Srivastava, Polly Ruhnke, Lavanya Vaddavalli, Sri Keshav Katragadda, Shweta Yadav, Md Shad Akhtar
January 22, 2025 at 6:31 AM
2/n This work is with
Authors: Abdullah Mazhar, Zuhair Hasan Shaik, Aseem Srivastava, Polly Ruhnke, Lavanya Vaddavalli, Sri Keshav Katragadda, Shweta Yadav, Md Shad Akhtar
Authors: Abdullah Mazhar, Zuhair Hasan Shaik, Aseem Srivastava, Polly Ruhnke, Lavanya Vaddavalli, Sri Keshav Katragadda, Shweta Yadav, Md Shad Akhtar