About
Research & Development
The Evolving Logic of Language Models: A Cost-Benefit Analysis
TLDR: A longitudinal study evaluated the logical reasoning of frontier Large Language Models (LLMs) from 2023 to 2025, showing performance improvements attributed to hidden...
Research & Development
Enhancing LLM Reasoning with Latent Thought Optimization
TLDR: LTA-Thinker is a new training framework for Large Language Models (LLMs) that improves complex reasoning by optimizing "Latent Thoughts." It uses a lightweight...
Research & Development
The Unpredictable Street: LLMs and the Challenge of Random Address Generation
TLDR: A study investigated Large Language Models' (LLMs) ability to generate random street addresses in Amsterdam, Madrid, Paris, and Rome. It found that LLMs...
Research & Development
HLSMAC: Advancing AI Strategy in StarCraft II with Ancient Wisdom
TLDR: HLSMAC is a new StarCraft II benchmark for multi-agent reinforcement learning that focuses on high-level strategic decision-making, moving beyond micromanagement. It features 12...
Research & Development
Evo-Merging: Combining Black-Box Language Models for Enhanced AI Services
TLDR: Evo-Merging is a new framework that allows for the merging of multiple large language models (LLMs) even when their internal parameters are inaccessible,...
Research & Development
Unpacking the Architecture of Success: How Text Structure Shapes Impact Across Media
TLDR: A study by Honna, Murayama, and Matsui investigates how text structure, defined by the order and timing of functional blocks, correlates with success...