HKBU NLP Lab
@HKBU_NLP
Followers
58
Following
25
Media
10
Statuses
21
🧠NLP Research Lab @hkbaptistu 🔍Large Language Models, Social Media Analysis, and more. 🌐Connect with us: [email protected]
Hong Kong SAR
Joined September 2023
Our work has been accepted to ACM Multimedia 2025! Can't wait to discuss GUI Agents in Dublin, Ireland.
🚀 Introducing ScreenSpot-Pro – the first benchmark driving Multi-modal LLMs into high-resolution professional GUI-Agent and computer-use environments!. 📊 While GUI agents excel at general tasks like web browsing, professional applications remain underexplored. 🔹 ScreenSpot-Pro
0
0
2
🥳 7 papers from our group have been accepted by ACL 2025. 🥳 4 Main Papers + 3 Finding Papers!.🥳 Amazing works by our students!
0
0
2
Amazing work by our PhD students and collaborators!.
🚀 Introducing ScreenSpot-Pro – the first benchmark driving Multi-modal LLMs into high-resolution professional GUI-Agent and computer-use environments!. 📊 While GUI agents excel at general tasks like web browsing, professional applications remain underexplored. 🔹 ScreenSpot-Pro
0
0
2
🎉One paper of our group is accepted to #COLING2025! .👉CodeJudge-Eval: Can Large Language Models be Good Judges in Code Understanding? 📘A benchmark designed to assess LLMs' code understanding from the perspective of code judging rather than generation.
0
0
3
Amazing work led by our PhD student, Ziyang Luo @ChiYeung_Law!.
Introducing VideoAutoArena 🎥⚔️.An automated and scalable arena-style benchmark for video understanding! Its rankings align closely with human judgment, tackling complex, open-ended video analysis beyond multi-choice questions. Enter the battleground:
0
0
1
RT @knmnyn: 📢Please RT!. 🫵Hey #nlproc @aclmeeting, Wed 6 Nov, #ssnlp2024 features 🎶keynotes @jasonwei (@OpenAI), Bing Liu (@thisisUIC), Yue….
0
8
0
🎉 Three papers from our group and collaborators have been accepted by EMNLP 2024. Amazing work by our students!
0
1
2
Amazing work by our PhD students @danielhzlin @ChiYeung_Law !.
🔥Introducing our MFC-Bench🔥.🚀It is a comprehensive Multimodal Fact-Checking benchmark designed to evaluate Large Multimodal Models in terms of identifying factual inconsistencies and counterfactual scenarios. 👉Paper:
0
0
2
Paper1: Towards Explainable Harmful Meme Detection through Multimodal Debate between Large Language Models.
0
0
0
🎉Glad to see that two papers of our group are accepted to #WWW2024!. Paper1: Towards Explainable Harmful Meme Detection through Multimodal Debate between Large Language Models. Paper2: Explainable Fake News Detection With Large Language Model via Defense Among Competing Wisdom.
1
1
4
Thrilled to announce the acceptance of WizardCoder, the work of our third-year PhD student, Ziyang Luo, at ICLR 24! 🌟.
✨Excited to share that WizardCoder's paper has been accepted at ICLR 24! Can't wait to meet new and old friends in Vienna.
0
0
2
RT @WizardLM_AI: 🔥 Excited to release WizardCoder-33B-V1.1, the SOTA OSS Code LLM. 🥇79.9% pass@1 on HumanEval, surpasses GPT3.5-Turbo, Dee….
0
94
0
Our extensive experiments across a range of LMMs (e.g. GPT-4V) reveal that current models still exhibit a deficiency in safety awareness, showing insensitivity to various forms of implicit abuse.
0
0
1
🚀 Excited to share "GOAT-Bench: Safety Insights to Large Multimodal Models through Meme-Based Social Abuse". 📊GOAT-Bench evaluates LMMs' proficiency in gauging hatefulness, misogyny, offensiveness, sarcasm, and harmful meme content. 💻ProjectPage:
2
1
11