🎉Our paper "Knowledge Graphs Meet Multi-Modal Learning: A Comprehensive Survey is released: KG-driven #multimodal (KG4MM) learning and Multi-Modal #KnowledgeGraph (MM4KG) & #LLM 🧐55 Pages, 11 Tabs, 13 Figs, 619 Citations 🌐 https://t.co/UddnoH19as 📷 https://t.co/Rr6uIoyf90

🚀After over a year of hard work, we’re thrilled to share InstructCell alongside our paper: "A Multi-Modal AI Copilot for Single-Cell Analysis with Instruction Following." #AI #LLM #SingleCellAnalysis #AI4Science #NLP✨ 🧬 InstructCell bridges natural language and gene

@YinFang22900365 @ZhangEach_ZJUKG @GengYuxia @wencolani @ChenJiaoyan1 @jpansw @zxlzr @ChenHuajun

ChatCell Facilitating Single-Cell Analysis with Natural Language paper page: https://t.co/X0hb1gt1rn As Large Language Models (LLMs) rapidly evolve, their influence in science is becoming increasingly prominent. The emerging capabilities of LLMs in task generalization and

Welcome to recommend missing papers through Adding Issues or Pull Requests in our Repo:

🔍 We focus on research in two principal aspects: KG-driven #multimodal (KG4MM) learning, where KGs support multi-modal tasks, and Multi-Modal #KnowledgeGraph (MM4KG), which extends KG studies into the #MMKG realm.

🎉 Our paper "Structure-CLIP: Towards Scene Graph Knowledge to Enhance Multi-modal Structured Representations" is accepted by #AAAI2024 🏝️#SceneGraph #Knowledge enhances #multimodal structure representation. 🧰 Paper: https://t.co/u2y5fkokp8 🔎Code: https://t.co/DUFizqeRcS

🔗 Github: https://t.co/1fgOG4ye05 🔍 What can LLMs do for KGs? Or, in other words, what role can KG play in the era of LLMs? 🙌 This repository collects papers integrating KGs and LLMs. 😎 Welcome to recommend missing papers. #KnowledgeGraph & #NLP & #LLM & #PaperList

🎉 Our paper "Knowledgeable Preference Alignment for LLMs in Domain-specific Question Answering" is released 📷 New paradigm for LLM to inject the knowledge 🌐Paper: https://t.co/TkjX2pHYAz 📊Code: https://t.co/8frmx9qyzO #KnowledgeGraph & #NLP & #LLM & #Alignment

@jpansw

@iswc_conf 🥳🥳🥳

🧵Propose our model UMAEA: • Multi-scale modality hybrid and circularly missing modality imagination • Consistently achieve SOTA results across all benchmark splits • Limited parameters and runtime

🧵Identify two critical phenomena: • Models may succumb to overfitting noise during training • Models exhibit performance oscillations or even declines at high missing modality rates

🧵Benchmark latest MMEA models: Standard (non-iterative) and iterative training paradigms

🧵Propose the MMEA-UMVM dataset: Contain seven separate datasets with a total of 97 splits

@YinFang22900365 @wencolani @ChenJiaoyan1 @ChenHuajun

🎉 We are thrilled to announce our paper "Rethinking Uncertainly Missing and Ambiguous Visual Modality in Multi-Modal Entity Alignment" (🥳 best paper nominated at @iswc_conf ) 🔬Code: https://t.co/NZdm7Njko5 🧐 Paper: https://t.co/wHoWTm0xiW #ISWC #EntityAlignment #MMKG

🌟 Excited to announce that Mol-Instructions has just been updated with train/test data splits and evaluation code! 🚀 Dive in and explore! 🛠️💡#NLP #LLM #MolecularAI #Instructions #Dataset 🌐 Code: https://t.co/2DljZhU4kD 📊 Data: https://t.co/HG3jdzbWXv

Thrilled to announce the release of our 7B molecular language model based on SELFIES! 🎉 Download: https://t.co/74mNDVvNQ1 Dive in to generate molecules from scratch using the bos_token or input a partial structure for completion. Have a try! 🔬 #MolecularAI #Innovation #NLP #AI

🎉 Our paper "Making Large Language Models Perform Better in Knowledge Graph Completion" is released. 💡 Enhance the effectiveness of LLMs in handling KGC. Paper: https://t.co/Rfw7NSSFCq Github: https://t.co/6eFzS1MEPO 🔬 #AI #KnowledgeGraph #LanguageModels #NLP