Sources
Loading...
Additional media
Loading...
DeepNewz, mobile.
People-sourced. AI-powered. Unbiased News.
The release of Prometheus 2, a state-of-the-art open-source language model (LM) with configurations of 7B & 8x7B, designed to evaluate other LMs, marks a significant advancement in the field of artificial intelligence. Developed to address the limitations of closed-source LMs like GPT-4, Prometheus 2 offers enhanced transparency, controllability, and affordability. It supports both direct assessments and pairwise ranking, allowing evaluations based on user-defined criteria. This development reflects ongoing discussions in the AI community about the challenges of model evaluation, highlighted by recent publications such as those from RekaAILabs and contributions by researchers like Seungone Kim.
[CL] Prometheus 2: An Open Source Language Model Specialized in Evaluating Other Language Models https://t.co/w9YNga3hir - Proprietary LMs such as GPT-4 are often used to evaluate other LMs, but open LMs are needed due to transparency, controllability, and affordability… https://t.co/dZMNsNyubY
🚨 New paper + models! Evaluating LLMs using closed-source LLMs has limited transparency, controllability, and affordability. Incredible work by @seungonekim significantly improves all these factors, w/ open models for either relative or absolute response scoring. ⬇️ https://t.co/RBVdas3dAb
#NLProc Introducing 🔥Prometheus 2, an open-source LM specialized on evaluating other language models. ✅Supports both direct assessment & pairwise ranking. ✅ Improved evaluation capabilities compared to its predecessor. ✅Can assess based on user-defined evaluation criteria. https://t.co/qsN8DG1l8L
