Mingda Chen

Last one of the year - EWE: arxiv.org/pdf/2412.18069 Ewe (Explicit Working Memory), enhances factuality in long-form text generation by integrating a working memory that receives real-time feedback from external resources.

Meta presents Improving Factuality with Explicit Working Memory Presents EWE, a novel approach that enhances factuality in long-form text generation by integrating a working memory that receives real-time feedback from external resources EWE outperforms strong baselines on four fact-seeking long-form generation datasets, increasing the factuality metric, VeriScore, by 2 to 10 points absolute without sacrificing the helpfulness of the responses. arxiv.org/abs/2412.18069

Newly published work from FAIR, Chameleon: Mixed-Modal Early-Fusion Foundation Models. This research presents a family of early-fusion token-based mixed-modal models capable of understanding & generating images & text in any arbitrary sequence. Paper ➡️ go.fb.me/7rb19n

Preview of our poster: #ICLR2024 #RAG #LLMs

🌐 Retrieval-augmentation is becoming increasingly popular for enriching LLMs with long-tail knowledge and keeping them up-to-date. 💫 Introducing RA-DIT: Retrieval-Augmented Dual Instruction Tuning (arxiv.org/abs/2310.01352), a fine-tuning process that enhances LM and retrieval integration in RALM, leading to better performance on knowledge-rich downstream tasks.

Yep, I'm back in Phantom #BruisesRus

SMH