Blog

Can Minor Document Typos Comprehensively Disrupt RAG Retriever & Reader Components? Retrieval-Augmented Generation (RAG) is effective in leveraging LLM in-context learning (ICL) capabilities.

Data Design For Fine-Tuning To Improve Small Language Model Behaviour Teaching Small Language Models to Self-Correct & Reason by using creative data formats for fine-tuning data. Via Prompt Erasure & Partial Answer Masking.

For Conversational User Interfaces to improve, the reliance on social intelligence will need to be leveraged. Many now regards social intelligence is a prerequisite for human-like Artificial Intelligence.

Agentic Search-Augmented Factuality Evaluator (SAFE) For LLMs This study demonstrates that an LLM agent can outperform crowdsourced human annotators…

There has been instances where another Language Model is used to vet the RAG output, which fails to detect incorrect and incomplete generated data.

Disambiguation: Using Dynamic Context In Crafting Effective RAG Question Suggestions This approach reminds of a technique initially implemented by IBM Watson called disambiguation. Where ambiguous input from the user is responded to with about five or less options to choose from. Hence allowing the user to perform disambiguation themselves, and allowing the system to learn from it.