Are We Ready for Multi-Image Reasoning? Launching VHs: The Visual Haystacks Benchmark!
TheStaff Humans excel at processing vast arrays of visual information, a skill that is crucial for...
TheStaff
How to Evaluate Jailbreak Methods: A Case Study with the StrongREJECT Benchmark
When we began studying jailbreak evaluations, we found a fascinating paper claiming that you could...
Linguistic Bias in ChatGPT: Language Models Reinforce Dialect Discrimination
Sample language model responses to different varieties of English and native speaker reactions. ChatGPT does...
Virtual Personas for Language Models via an Anthology of Backstories
We introduce Anthology, a method for conditioning LLMs to representative, consistent, and diverse virtual personas...
π0 Released and Open Sourced: A General-Purpose Robotic Foundation Model that could be Fine-Tuned to a Diverse Range of Tasks
Robots are usually unsuitable for altering different tasks and environments. General-purpose models of robots are...
Prime Intellect Releases SYNTHETIC-1: An Open-Source Dataset Consisting of 1.4M Curated Tasks Spanning Math, Coding, Software Engineering, STEM, and Synthetic Code Understanding
In artificial intelligence and machine learning, high-quality datasets play a crucial role in developing accurate...
Chain-of-Associated-Thoughts (CoAT): An AI Framework to Enhance LLM Reasoning
Large language models (LLMs) have revolutionized artificial intelligence by demonstrating remarkable capabilities in text generation...
Weaviate Researchers Introduce Function Calling for LLMs: Eliminating SQL Dependency to Improve Database Querying Accuracy and Efficiency
Databases are essential for storing and retrieving structured data supporting business intelligence, research, and enterprise...
Princeton University Researchers Introduce Self-MoA and Self-MoA-Seq: Optimizing LLM Performance with Single-Model Ensembles
Large Language Models (LLMs) such as GPT, Gemini, and Claude utilize vast training datasets and...
Optimizing Large Model Inference with Ladder Residual: Enhancing Tensor Parallelism through Communication-Computing Overlap
LLM inference is highly resource-intensive, requiring substantial memory and computational power. To address this, various...