[1/n]
🎉🎉🎉 Excited to share our latest work: 'The Fine Line: Navigating Large Language Model Pretraining with Down-streaming Capability Analysis'! We delve into the dynamics of LLMs across different scales and domains.

💡Highlights include:

🗺️ Comprehensive Model Evaluation:

[1/n]
🚀 Excited to share our latest work on OpenCodeInterpreter! With a blend of execution results and human feedback, we've achieved significant advancements in code generation. Here are the key points:

✨ Introducing OpenCodeInterpreter - a leap in iterative code refinement.

Nice!😺

MathPile: A Billion-Token-Scale Pretraining Corpus for Math

paper page: huggingface.co/papers/2312.17…

High-quality, large-scale corpora are the cornerstone of building foundation models. In this work, we introduce MathPile, a diverse and high-quality math-centric corpus comprising

A research on the reliability of Large Language Models' behavior #LLM #LLM s #ChatGPT . I've encountered similar situations in practical use, which left me puzzled about the LLMs generating different results🤨.

Welcome to check our #sigir2023 paper🥳
dl.acm.org/doi/10.1145/35…

When we started this project, ChatGPT had not yet appeared. This paper may not have been so attractive in the ChatGPT era. But I still want to share our insights from this project, in hopes more people will learn

Multimodal Emotion Recognition in Multiparty Conversations (MERMC) currently focus on text and audio modalities. How do we extract facial sequences of real speaker and further help MER? Introducing  Facial expression-aware Multimodal Multi-Task learning (FacialMMT) #acl2023

Interesting!

I think this in-depth analysis of fake news is also highly valuable in addressing the issue of hallucination in large language models. It provides valuable insights into potential solutions for this problem.