AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

Query-Level Uncertainty in Large Language Models

Lihu Chen, Gerard de Melo, Fabian M. Suchanek, Gaël Varoquaux

ICLR 2026 · Large language models › Retrieval and knowledge · approach: Long-context and personalized retrieval · 10 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 P(Yes) over layers and tokens with attenuated weights instead of naive averaging (ablation)Internal Confidence · TriviaQA/SciQ/GSM8K averageAUROC: 57.4 -> 59.8 (Phi), 61.7 -> 64.2 (Llama), 63.5 -> 67.1 (Qwen)Improved
aggregate P(Yes) around the decision center instead of using only the last layer and last token (ablation)Internal Confidence · TriviaQA/SciQ/GSM8K averageAUROC: 59.8 -> 59.8 (Phi), 55.5 -> 64.2 (Llama), 60.9 -> 67.1 (Qwen)Mixed
use Internal Confidence instead of Predictive Entropy on GSM8Kquery-level uncertainty · GSM8KAUROC: 63.6 vs 53.9 (Phi), 66.1 vs 65.7 (Llama), 66.4 vs 66.8 (Qwen)Mixed
use Internal Confidence instead of answer-level MSPuncertainty estimation · GSM8K (Qwen-14B)AUROC: 68.5 -> 66.8Hurt
use query-level Internal Confidence instead of answer-level SARuncertainty estimation · GSM8K (Qwen-14B)time per sample (s): 180.6 -> 0.3Improved
use query-level Internal Confidence instead of answer-level methodsuncertainty estimation · TriviaQA (Qwen-14B)AUROC: 78.1 (EV Laplacian) -> 71.9Hurt
use Internal Confidence instead of Perplexityuncertainty estimation · SciQ (Qwen-14B)AUROC: 71.5 -> 62.6Hurt
use Internal Confidence instead of P(Yes) top rightquery-level uncertainty · SimpleQA/MuSiQue/TruthfulQA averageAUROC: 55.4 -> 61.0 (Phi), 53.7 -> 61.0 (Llama), 52.0 -> 57.2 (Qwen)Improved
use Internal Confidence instead of P(Yes) naive avgquery-level uncertainty · TriviaQA ECE (Llama-8B)ECE: 12.6 -> 25.4Hurt
use Internal Confidence instead of P(Yes) naive avgquery-level uncertainty · SimpleQA/MuSiQue/TruthfulQA average ECE (Llama-8B)ECE: 16.8 -> 52.4Hurt
vary locality alpha (default 1.0, locality about 0.7) (ablation)Internal Confidence · TriviaQA/SciQ/GSM8KAUROCMixed
add in-context examples (1 to 5 shots) (ablation)Internal Confidence · TriviaQA/SciQ/GSM8KAUROCNo clear effect

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