AI Paper Atlas

Research map / Large language models / Alignment and preferences

Query-Routed Activation Editing with Truth-hierarchical Preference Optimization

Kewei Liao, Tianbo Wang, Yuqing Ma, Zhange Zhang, Zhicheng Geng, Xiaowei Zhao, Jiakai Wang, Xianglong Liu

AAAI 2026 · Large language models › Alignment and preferences · approach: Active and token-level preference learning · 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
add DHR query-routed head selection (ablation)no-editing baseline · TruthfulQA (open-ended)True*Info: 62.0 -> 74.2Improved
add TPS preference steering (ablation)no-editing baseline · TruthfulQA (open-ended)True*Info: 62.0 -> 83.3Improved
add TPS on top of DHR (ablation)QRAE · TruthfulQA (open-ended)True*Info: 74.2 -> 85.1Improved
add DHR on top of TPS (ablation)QRAE · TruthfulQA (multiple-choice)MC1: 61.3 -> 63.3Improved
add untruthful-to-truthful loss to DHR (ablation)QRAE TPS · TruthfulQA (open-ended)True*Info: 74.2 -> 81.3Improved
add truthful-to-best loss to DHR (ablation)QRAE TPS · TruthfulQA (open-ended)True*Info: 74.2 -> 78.5Improved
add truthful-to-best loss on top of untruthful-to-truthful loss (ablation)QRAE TPS · TruthfulQA (open-ended)True*Info: 81.3 -> 85.1Improved
use DHR head selection instead of ITI probe-accuracy selection (ablation)global editing vector · TruthfulQA (open-ended)True*Info: 69.0 -> 74.2Improved
use DHR head selection instead of random selection (ablation)global editing vector · TruthfulQA (open-ended)True*Info: 65.8 -> 74.2Improved
use DHR head selection instead of bias-based scoring (ablation)global editing vector · TruthfulQA (open-ended)True*Info: 52.2 -> 74.2Improved
increase edited heads K from 16 to 48 (ablation)QRAE · TruthfulQA (open-ended)True*Info: 77.5 -> 85.1Improved
increase edited heads K beyond 48 up to 96 (ablation)QRAE · TruthfulQA (open-ended)True*Info / MC1No 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