About
Research & Development
Boosting E-commerce Catalog Quality: A New LLM Prompting System
TLDR: A novel, training-free system utilizes a cascade of Large Language Models (LLMs) to automatically generate and refine prompts for assessing product quality in...
Research & Development
Assessing Large Language Models’ Chess Understanding with ChessQA
TLDR: ChessQA is a new, comprehensive benchmark designed to evaluate how well large language models (LLMs) truly understand chess, moving beyond simple move quality....
Research & Development
SAFEVISION: A New Standard for Image Content Moderation
TLDR: SAFEVISION is a novel image guardrail system that combines human-like reasoning with efficient automation to detect and explain unsafe content. It features a...
Research & Development
Improving LLM Reliability Through Semantic Confidence Rewards
TLDR: This research introduces FISCORE, a novel reinforcement learning framework designed to enhance the reliability of Large Language Models (LLMs) by teaching them to...
Research & Development
MGAudio: A Self-Guiding Framework for Advanced Video-to-Audio Generation
TLDR: MGAudio is a new flow-based framework for open-domain video-to-audio generation that uses a novel model-guided dual-role alignment approach. Unlike previous methods relying on...
News & Current Events
GitHub Introduces Agent HQ: A Unified Platform for AI-Powered Coding Collaboration
TLDR: GitHub has launched 'Agent HQ,' a new platform designed to integrate AI coding agents from leading providers like OpenAI, Google, Anthropic, Cognition, and...