The #G20, a loose group of economically-powerful countries, this weekend announced that it was welcoming the #AfricanUnion (#AU) as a new member. https://www.euronews.com/2023/09/09/african-union-to-become-permanent-member-of-g20-forum

One AU member country, #SouthAfrica, was already part of the G20, and will presumably participate separately.

#intergovernmental #geography #AfricanEconomics #Africa #world #internationalrelations #IR @geography @geopolitics

📝 PRISTA-Net: Deep Iterative Shrinkage Thresholding Network for Coded Diffraction Patterns Phase Retrieval 🔭💿

"A learnable nonlinear transformation to address the proximal-point mapping sub-problem, and an attention mechanism to focus on phase information, global feature learning via FFT, and log-based loss function." [gal30b+] 🤖 #CV #IR #IT

⚙️ https://github.com/liuaxou/PRISTA-Net
🔗 https://arxiv.org/abs/2309.04171v1 #arxiv

📝 Extending Transductive Knowledge Graph Embedding Models for Inductive Logical Relational Inference 👾💿

"Extends transductive models via a harmonic extension which leverages entity representations learned via transductive embedding methods to infer entity representations of new entities introduced at inference time as in the inductive setting." [gal30b+] 🤖 #AI #IR #SI

⚙️ https://github.com/tgebhart/
🔗 https://arxiv.org/abs/2309.03773v1 #arxiv

From Kosovo to Ukraine

https://www.europe-solidaire.org/spip.php?article67600

Gilbert "Achcar’s The New Cold War deftly arms internationalists with the historical knowledge needed to develop new strategies to build the international working class in a new period of inter-imperialist rivalry. What these strategies may be, however, remains only to be discovered beyond the pages of The New Cold War—in the existing conjuncture of struggle as it develops day by day".

#Marxism #internationalism #IR #conflict #NATO #Imperialism

📝 Impression-Informed Multi-Behavior Recommender System: A Hierarchical Graph Attention Approach 💿🧠

"Hierarchical Multi-behavior Graph Attention Network (HMGN) uses a specialized multi-behavior sub-graph sampling technique to integrate a hierarchical attention mechanism that captures information from both inter and intra-behaviors." [gal30b+] 🤖 #IR #LG

🔗 https://arxiv.org/abs/2309.03169v1 #arxiv

📝 Prompt-Based Effective Input Reformulation for Legal Case Retrieval 💿

"PromptCase first identifies the key features for legal case retrieval, legal facts and legal issues, and then leverages a prompt-based encoding scheme to conduct effective legal case retrieval." [gal30b+] 🤖 #IR

⚙️ https://github.com/yanran-tang/PromptCase
🔗 https://arxiv.org/abs/2309.02962v1 #arxiv

📝 Tidying Up the Conversational Recommender Systems' Biases 💿

"Provides a concise review of biases in conversational recommender systems by surveying recent literature on recommender systems, conversational systems, and language models in the context of conversational recommendation." [gal30b+] 🤖 #IR

🔗 https://arxiv.org/abs/2309.02550v1 #arxiv

📝 Towards Contrastive Learning in Music Video Domain 💿🔭🔊

"Contrastive learning is a powerful way of learning multimodal representations across various domains such as image-caption retrieval and audio-visual representation learning but it might be not the best choice of learning audio-visual representations for music videos." [gal30b+] 🤖 #IR #CV #MM #SD

⚙️ https://github.com/KarelVeldkamp/Multimodal-Musicvideo-Representation
🔗 https://arxiv.org/abs/2309.00347v1 #arxiv

📝 Co-Evolving Vector Quantization for ID-based Recommendation 💿

"Develops and integrate a novel co-evolving vector quantization framework, namely COVE, into various recommendation models and tasks including list completion, collaborative filtering, and click-through rate prediction." [gal30b+] 🤖 #IR

⚙️ https://github.com/Jyonn/Cove
🔗 https://arxiv.org/abs/2308.16761v1 #arxiv

📝 Context Aware Query Rewriting for Text Rankers Using LLM 💿👾

"A ranker is fine-tuned on the rewritten queries instead of the original queries during training and inference phase, using only relevant documents as context to generate re-writes." [gal30b+] 🤖 #IR #AI

🔗 https://arxiv.org/abs/2308.16753v1 #arxiv

📝 AntM$^{2}$C: A Large Scale Dataset for Multi-Scenario Multi-Modal CTR Prediction 💿🧠

"AntM$^{2}$C is a multi-scenario multi-modal CTR dataset which includes data about 200 million users and 6 million items in total." [gal30b+] 🤖 #IR #LG

🔗 https://arxiv.org/abs/2308.16437v1 #arxiv

📝 Adaptive Multi-Modalities Fusion in Sequential Recommendation Systems 💿

"MMSR represents each user's history as a graph, and incorporates dual attention to differentiate homogeneous and heterogeneous neighbors during graph propagation, enabling each modality to prioritize either its inherent sequential nature or its interplay with other modalities." [gal30b+] 🤖 #IR

🔗 https://arxiv.org/abs/2308.15980v1 #arxiv

📝 Denoising Attention for Query-Aware User Modeling in Personalized Search 💿

"Denoising Attention mechanism is based on two novelties, namely, Soft-Denoising Attention and Hard-Denoising Attention: the first one is based on the Softmax normalization and the second on the Hardmax normalization." [gal30b+] 🤖 #IR

⚙️ https://github.com/google/cld3
🔗 https://arxiv.org/abs/2308.15968v1 #arxiv

📝 DRGame: Diversified Recommendation for Multi-Category Video Games with Balanced Implicit Preferences 💿

"Centered on multi-category video games, consisting of two components: Balance-driven Implicit Preferences Learning for data pre-processing and Clustering-based Diversified Recommendation Module for final prediction." [gal30b+] 🤖 #IR

🔗 https://arxiv.org/abs/2308.15823v1 #arxiv

📝 A Survey on Multi-Behavior Sequential Recommendation 💿

"Set up to address the issue of information overload in traditional information retrieval systems, which is focused on recommending information that is of most interest to users from massive information." [gal30b+] 🤖 #IR

🔗 https://arxiv.org/abs/2308.15701v1 #arxiv

📝 Chunked Lists Versus Extensible Arrays for Text Inversion 💿

"In both cases the data is split into small blocks and indexed by an in-memory array (FBB) or a disk-based B+ tree (SQA)." [gal30b+] 🤖 #IR

🔗 https://arxiv.org/abs/2308.15498v1 #arxiv

📝 Knowledge-Grounded Natural Language Recommendation Explanation 📚💿

"Proposes a KG approach to natural language explainable recommendation, which draws on user-item features through a novel collaborative filtering-based KG representation to produce fact-grounded, personalized explanations, while jointly learning user-item representations for recommendation scoring." [gal30b+] 🤖 #CL #IR

🔗 https://arxiv.org/abs/2308.15813v1 #arxiv

📝 Robust Long-Tailed Learning via Label-Aware Bounded CVaR 🧠🔭💿

"Proposes a LAB-CVaR with logit adjustment (LAB-CVaR-logit) loss to optimize the model and improve the performance of long-tailed learning with a solid theoretical ground." [gal30b+] 🤖 #LG #CV #IR

🔗 https://arxiv.org/abs/2308.15405v1 #arxiv

📝 Classification-Aware Neural Topic Model Combined with Interpretable Analysis -- For Conflict Classification 🧠📚💿

"An improved classification-aware neural topic model for conflict information classification and topic discovery in social media posts, which introduces the interpretation into the model architecture to improve the performance and interpretation quality of the topic model." [gal30b+] 🤖 #LG #CL #IR

🔗 https://arxiv.org/abs/2308.15232v1 #arxiv

📝 A Multi-Perspective Learning to Rank Approach to Support Children's Information Seeking in the Classroom 💿

"Introduces a novel re-ranking model that aims to augment the functionality of standard search engines to support classroom search activities for children (ages 6 to 11)." [gal30b+] 🤖 #IR

⚙️ https://github.com/Neelik/REdORank
🔗 https://arxiv.org/abs/2308.15265v1 #arxiv