Topic

Natural language processing

Discover key takeaways from 2 podcast episodes about this topic.

Cascading Adversarial Bias from Injection to Distillation in Language Models
Large Language Models (LLMs)Retrieval Augmented Generation (RAG)Adversarial Attacks
Jan 27, 2026

Cascading Adversarial Bias from Injection to Distillation in Language Models

RAG systems, designed to enhance LLM accuracy and personalization, are vulnerable to 'Phantom' trigger attacks where a single poisoned document can manipulate outputs to deny service, express bias, exfiltrate data, or generate harmful content.

G
Google TechTalks
Data Mixture Inference: What do BPE Tokenizers Reveal about their Training Data?
LLM training dataBPE tokenizersData mixture inference
Jan 27, 2026

Data Mixture Inference: What do BPE Tokenizers Reveal about their Training Data?

BPE tokenizers, often overlooked, provide a transparent and accessible window into the secret data mixtures used to train large language models.

G
Google TechTalks