Research & Innovation

Global: Cross-User KV Sharing Cuts Cache Size to 0.8% in Sequential Recommendation Models A new method named CollectiveKV has been introduced to address latency and storage concerns in sequential recommendation…

Global: Jenius-Agent Framework Boosts Reliability of Long-Horizon Tool-Augmented LLM Agents Background and Motivation Researchers at Jenius and collaborating institutions have introduced a new agent framework designed to address persistent challenges…

Global: Multi-Agent System Boosts Accuracy of Procedural Graph Extraction from TextA team of computer scientists has introduced a new multi‑agent framework designed to extract procedural workflows as graphs from natural‑language…

Global: Researchers Unveil Hierarchical Localization Agent That Improves Image Geolocation AccuracyResearchers at an international team have introduced a hierarchical localization system called LocationAgent that improves image geolocation performance, particularly in…

Global: Researchers Use LLMs to Create Diverse Black‑Box Optimization BenchmarksOn January 31, 2026, a team of researchers posted a new study on arXiv that details an approach for generating continuous…

Global: New Multi-Agent Framework Boosts Zero-Shot Time Series Reasoning Researchers have introduced TS-Debate, a modality‑specialized, collaborative multi‑agent debate system designed to improve zero‑shot reasoning on time‑series data. The framework was…

Global: Diffusion-Driven Bayesian Exploration Framework Mitigates Early-State Estimation Bias in Emergency ResponseA team of researchers has introduced a diffusion-driven Bayesian exploration framework (DEPF) to correct early-stage state estimation errors that…

Global: Length-Adaptive Interest Network Improves CTR Models Across Varying User Sequence LengthsIn January 2026, researchers released a study on arXiv that introduces a Length‑Adaptive Interest Network (LAIN) designed to mitigate…

Global: Synthetic Object Compositions Pipeline Boosts Vision Model Performance Overview of the New Synthesis Approach Researchers have introduced a pipeline called Synthetic Object Compositions (SOC) that generates high‑quality synthetic images…

Global: Adaptive Neuro-Fuzzy System Offers Explainable Uncertainty for Wastewater Energy ForecastsResearchers at Melbourne Water have introduced an Interval Type-2 Adaptive Neuro-Fuzzy Inference System (IT2-ANFIS) to improve short-term electricity demand forecasting…