Why Multimodal LLMs Fail at Retrieval, Reasoning-Enhanced Sequential Modeling for Industrial Recommendation, and More!
Vol.136 for Dec 22 - Dec 28, 2025
Stay Ahead of the Curve with the Latest Advancements and Discoveries in Information Retrieval.
This week’s newsletter highlights the following research:
Why Multimodal LLMs Fail at Retrieval, from Feng et al.
Reasoning-Enhanced Sequential Modeling for Industrial Recommendation, from Alibaba
Transferring Reasoning Strategies to Lightweight Models in RAG Systems, from Chen et al.
Efficient Optimization of Hierarchical Identifiers for Generative Recommendation, from the University of Amsterdam
Lightweight Solutions for Cold-Start Item Prediction in Production Recommender Systems, from Pinterest
Automatic Multimodal Knowledge Graph Construction for Enhanced RAG, from Hsiao et al.
Efficient Dense Retrievers Through MLP-Focused Compression of Large Language Models, from Lei et al.
A Closed-Loop Approach to Memory Retrieval in LLM Agents, from MBZUAI
Enhancing Few-Shot Learning Through Retrieval-Augmented Prompting, from Chen et al.
Measuring Retrieval Effectiveness When Total Relevant Documents Are Unknown, from PrimerAI
