4 papers
Semantic Faithfulness and Entropy Production Measures to Tame Your LLM Demons and Manage Hallucinations
Igor Halperin
Evaluating faithfulness of Large Language Models (LLMs) to a given task is a complex challenge. We propose two new unsupervised metrics for faithfulness evaluation using insights f…
Topic Identification in LLM Input-Output Pairs through the Lens of Information Bottleneck
Igor Halperin
Large Language Models (LLMs) are prone to critical failure modes, including \textit{intrinsic faithfulness hallucinations} (also known as confabulations), where a response deviates…
Prompt-Response Semantic Divergence Metrics for Faithfulness Hallucination and Misalignment Detection in Large Language Models
Igor Halperin
The proliferation of Large Language Models (LLMs) is challenged by hallucinations, critical failure modes where models generate non-factual, nonsensical or unfaithful text. This pa…
CAISSON: Concept-Augmented Inference Suite of Self-Organizing Neural Networks
Igor Halperin
We present CAISSON, a novel hierarchical approach to Retrieval-Augmented Generation (RAG) that transforms traditional single-vector search into a multi-view clustering framework. A…