🏃 Zhepei Wei presented "Fast and Accurate Language Model Decoding via Parallel Token Processing"

openreview.net/forum?id=coZ...

@zhepei.bsky.social @wlchen.bsky.social, Xinyu Zhu, Yu Meng

#NeurIPS2024 #AdaptiveFoundationModels

🥳 Finally arrived in Vancouver! Thrilled to be at #NeurIPS2024 this week, presenting our latest works on #LLMs:
1️⃣ ParaDecode (oral @ 2:10 pm)
2️⃣ InstructRAG (poster @ 4:30 pm)

Find me at Exhibit Hall A on Dec. 14—always happy to meet old & new friends and chat about anything! 😉

InstructRAG: Instructing Retrieval-Augmented Generation via Self-Synthesized Rationales Retrieval-augmented generation (RAG) has shown promising potential to enhance the accuracy and factuality of language models (LMs). However, imperfect retrievers or noisy corpora can introduce mislead...

Great list! Thanks for the efforts, Orion! Our recent work enabled LLMs to generate their OWN supervision for denoising retrieved information in RAG, achieving better accuracy and enhanced verifiability. If you’re interested, feel free to check out InstructRAG! paper link: arxiv.org/abs/2406.13629

🙌