AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Does Question Really Matter? The Attribution of Answer Bias in LLM Evaluation

Boxi Cao, Ruotong Pan, Hongyu Lin, Xianpei Han, Le Sun

AAAI 2026 · Large language models › Retrieval and knowledge · approach: Knowledge-augmented question answering · 1 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
shuffle option order in option-only evaluation (ablation)17 LLMs · ARC-challenge option-onlyaccuracy: 38.37% -> 38.19%No clear effect
replace wrong options with random wrong options from other questions (ablation)Qwen2-72B · MMLU option-onlyaccuracy: 48.8% -> 35.3%Hurt
train on test data (more contaminated samples) (ablation)base LLMs · MMLU/CMMLU/HellaSwag option-onlyoption-only accuracy: Pearson r 0.935, 0.991, 0.925Improved
train on contaminated subjects vs uncontaminated control subjects (ablation)LLaMA-3-8B · CMMLUoption-only accuracy: contaminated 62.9 vs control 34.4Improved
use OPD instead of BLD for contamination detectionLLaMA-2-7B contaminated with HellaSwag · contamination detectionF1: 0.558 -> 0.802Improved
raise OPD threshold T from 2 to 4 (ablation)OPD · contamination detectionF1: precision 0.764 -> 0.921; F1 0.802 -> 0.714Mixed
prune OPD-flagged instances from benchmark17 LLMs · HellaSwagaccuracy: average -22.77%Hurt

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