8 January 2025 Archives

machine learning Research

Optimizing LLM Test-Time Compute Involves Solving a Meta-RL Problem

January 8, 2025

Figure 1: Training models to optimize test-time compute and learn “how to discover” correct responses, as opposed to the traditional learning paradigm of learning “what answer” to output. The major strategy to improve large language models (LLMs) thus far has…

1.21K 51423

Machine Learning Blog | ML@CMU | Carnegie Mellon University

Statistics:

Categories:

January 8, 2025