Scaling LLM Test-Time Compute Optimally Can Be More Effective than Scaling Model Parameters
Snell et al., 2025, Scaling LLM Test-Time Compute Optimally Can Be More Effective than Scaling Model Parameters (The Thirteenth International Conference on Learning Representations (ICLR 2025))
Subscriber deck — free during early access
1 cards in “3.1 Test-Time Compute-Optimal Scaling Strategy” · whole deck
3.1 Test-Time Compute-Optimal Scaling Strategy
- What is the test-time compute-optimal scaling strategy?For a prompt \(q\) and budget \(N\), choose the hyperparameters \(\theta\) of the test-time method (e.g. revisions vs parallel samples, beam width) that maximize the probability of the correct answer: \(\theta^*_q(N) = \arg\max_\theta \mathbb{E}_{y \sim \mathrm{Target}(\theta, N, q)}[\mathbb{1}_{y = y^*(q)}]\).
- Snell et al., 2025, Scaling LLM Test-Time Compute Optimally Can Be More Effective than Scaling Model Parameters (The Thirteenth International Conference on Learning Representations (ICLR 2025)), §3.1
Source
- Charlie Snell, Jaehoon Lee, Kelvin Xu and Aviral Kumar. 2025. Scaling LLM Test-Time Compute Optimally Can Be More Effective than Scaling Model Parameters. In The Thirteenth International Conference on Learning Representations (ICLR 2025). https://arxiv.org/abs/2408.03314