Reading List

Sep 2, 2025512

Use LLMs to act on provided facts, not as lossless sources of exact details.

AI Hallucinations Prompt Engineering Retrieval-Augmented Generation Human-AI Collaboration

Bandit-Based, Budget-Aware LLM Routing with Preference-Informed LinUCB (PILOT)

Sep 1, 2025206

Treat LLM routing as a contextual bandit and use a preference-informed LinUCB plus a knapsack budget policy to adaptively, cost-effectively pick the right model per query.

LLM Routing Reinforcement Learning AI Infrastructure Algorithms & Optimization

Under the Hood

The Dimensional Ceiling of Single-Vector Embedding Retrieval

Aug 30, 2025151

Embedding-based retrieval hits a hard top-k capacity ceiling set by embedding dimension, and real systems already run into it.

Vector Embeddings Information Retrieval AI Benchmarks Search Quality

Reading List

LLMs Are Lossy Encyclopedias: Give Them Facts to Work With

Bandit-Based, Budget-Aware LLM Routing with Preference-Informed LinUCB (PILOT)

The Dimensional Ceiling of Single-Vector Embedding Retrieval