AI Paper Atlas

Research map / Large language models / Reasoning and chain of thought

Truthfulness Does Not Scale Like Reasoning: Why Polling Fails as a Proxy Verifier

Yegor Denisov-Blanch, Joshua Kazdan, Jessica Chudnovsky, Rylan Schaeffer, Sheng Guan, Soji Adeshina, Sanmi Koyejo

ICML 2026 · Large language models › Reasoning and chain of thought · approach: Test-time compute scaling

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Examines whether scaling inference compute via polling (sampling multiple answers and aggregating) improves truthfulness in domains lacking external verifiers, finding it does not scale like reasoning gains.

Problem

Whether inference-time compute scaling (e.g., pass@k style sampling and aggregation) can elicit truthfulness gains in domains without convenient external verifiers, unlike math and code.

Approach

Tests polling-based inference scaling as a proxy verifier for truthfulness across five benchmarks and models, and argues that truthfulness does not scale like reasoning with more compute.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
aggregate samples by majority vote instead of a single sampleLLM ensembles · HLE, BoolQ, Com2Sense, Predict-the-FutureaccuracyNo clear effect
scale to 25x inference cost with polling aggregationLLMs · verifier-absent benchmarksaccuracyNo clear effect
weight votes by self-reported confidencemajority voting · HLEaccuracy (%): 5.7 -> 2.9Hurt
weight votes by self-reported confidencemajority voting · Com2Senseaccuracy (%): 85.0 -> 82.0Hurt
use highest-confidence answermajority voting · HLEaccuracy (%): 5.7 -> 11.4Improved
use Surprisingly Popular rulemajority voting · HLEaccuracy (%): 5.7 -> 20.0Improved
use Surprisingly Popular rulemajority voting · BoolQaccuracy (%): 77.0 -> 76.0Hurt
invert the SP signalSP aggregation · HLEaccuracy: 80%Improved
use inverse-SPSP aggregation · other datasetsaccuracyNo clear effect
aggregate on forecasting questions postdating knowledge cutoffsall aggregation methods · Predict-the-FutureaccuracyNo clear effect
raise temperature from 0.7 to 1.0 to induce diversity (ablation)LLM sampling · question-model pairsplurality answer flips: 2.9%No clear effect
pool across five models instead of one model (ablation)ensembles · MATHconvergence to same wrong answer: 53%Hurt

Surprising findings

Experimental setup

Builds on: pass@k

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