AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Optimal Bayesian Stopping for Efficient Inference of Consistent LLM Answers

Jingkai Huang, Will Ma, Zhengyuan Zhou

ICML 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Nonparametric and PAC-Bayes Bayesian learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
aggregate to L=3 instead of exact posterior (ablation)L-aggregated Bayesian stopping · Synthetic (K=8, 1-δ=0.99)number of generations: 18.13 -> 18.07No clear effect
use L=3 instead of coarsest L=2 (ablation)L-aggregated Bayesian stopping · Synthetic (K=8, 1-δ=0.99)number of generations: 22.43 -> 18.07Improved
increase L from 3 to 4 (ablation)L-aggregated Bayesian stopping · Synthetic (K=8)posterior computation time (ms): 14.2 -> 37.4Hurt
increase L from 3 to 4 (ablation)L-aggregated Bayesian stopping · Synthetic (K=8, 1-δ=0.99)number of generations: 18.07 -> 18.11No clear effect
use L=3 instead of L=2 at 1-δ=0.9 (ablation)L-aggregated Bayesian stopping · Synthetic (K=8, 1-δ=0.9)number of generations: 10.74 -> 10.12Improved
add known-prior Bayesian stopping (L=3) instead of prior-free ASCL-aggregated Bayesian stopping · Synthetic (K=8, 1-δ=0.9)number of generations: 16.72 -> 10.12Improved
use known-prior L=3 instead of ASCL-aggregated Bayesian stopping · CommonsenseQA, Qwen-2.5-72B (1-δ=0.95)number of generations: 5.28 -> 3.38Improved
use uncertain-prior L=3 instead of ASCL-aggregated Bayesian stopping · CommonsenseQA, Qwen-2.5-72B (1-δ=0.95)number of generations: 5.28 -> 1.00Improved
use uncertain-prior L=3 instead of ASCL-aggregated Bayesian stopping · CommonsenseQA, LLaMA-3.1-405B (1-δ=0.95)answer accuracy: 90.4% -> 88.6%Hurt
use uncertain-prior L=3 instead of ASCL-aggregated Bayesian stopping · CommonsenseQA, Qwen-2.5-72B (1-δ=0.95)answer accuracy: 87.2% -> 86.3%Hurt
use uncertain-prior L=3 instead of L=2 (ablation)L-aggregated Bayesian stopping · CommonsenseQA, LLaMA-3.1-405B (1-δ=0.95)number of generations: 5.63 -> 4.58Improved
use known-prior L=3 instead of L=2 (ablation)L-aggregated Bayesian stopping · CommonsenseQA, LLaMA-3.1-405B (1-δ=0.95)number of generations: 6.92 -> 6.50Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy