Atom of Thoughts (AOT) — A Markovian Take on LLM ReasoningLLMs have seen major improvements with training-time scaling — increasing parameters and training data boosts performance. But once…Mar 4Mar 4
AdaptiveStep — Smarter Stepwise Reasoning in LLMs with Confidence-Based DivisionOne of the key challenges in deploying large language models (LLMs) for reasoning-intensive tasks — whether in mathematics, coding, or…Mar 3Mar 3
ARMAP: Scaling Autonomous Agents via Automatic Reward Modeling and PlanningLLMs struggle with multi-step decision-making and real-world interaction. The ARMAP framework introduced by Chen et al. (2025) and…Feb 23Feb 23
Structured Retrieval Orchestration: Why Multi-Agent Systems Need More Than RAGThe Limitation of RAG as an Isolated SystemFeb 191Feb 191
I Read The Chief AI Officer’s Handbook So You Don’t Have To — Here’s What Actually MattersAI leadership is evolving, and The Chief AI Officer’s Handbook attempts to capture what it takes to lead AI initiatives successfully. But…Feb 19Feb 19
Why Trusting LLM Outputs in Production Can Be Misleading and How We Can Quantify ItWhy Should We Care About LLM Uncertainty?Feb 181Feb 181
MARCO: Multi-Agent Real-time Chat OrchestrationReal-world deployment of multi-agent LLM based automation still faces major hurdles — inconsistencies, hallucinations, and inefficient…Feb 17Feb 17
The Internet of Agents (IoA): Protocol for Autonomous AI CollaborationWhy Multi-Agent Systems Need a RethinkFeb 16Feb 16
DeepSeek-R1: RL for LLMs RethoughtFor years, supervised fine-tuning (SFT) and reinforcement learning from human feedback (RLHF) have been the dominant training methods for…Jan 311Jan 311
Reimagining Plugin Orchestration with Turn-Based Multi-Agent Reinforcement LearningModern AI systems often rely on orchestrating multiple interconnected components — plugins for data sourcing, analysis, and advanced…Jan 182Jan 182
