AI Paper Atlas

Research map / Large language models / Instruction and fine-tuning

Towards Steering without Sacrifice: Principled Training of Steering Vectors for Prompt-only Interventions

Yuntai Bao, Qinfeng Li, Xinyan Yu, Ge Su, Wenqi Zhang, Liu Yan, Haiqin Weng, Jianwei Yin, Xuhong Zhang

ICML 2026 · Large language models › Instruction and fine-tuning · approach: Fine-tuning generalization and delta compression · 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
train steering factor jointly with direction instead of fixed unit factor (ablation)FSSV (Lang.) · AxBench CONCEPT500overall steering score: 0.072 -> 0.736 (Gemma2-2B)Improved
add joint training instead of factor sampling plus selection (ablation)FSSV (Lang.) · AxBench CONCEPT500overall steering score: 0.663 -> 0.736 (Gemma2-2B)Improved
add joint training to SimPO FSSV (RePS) on Gemma2-9B (ablation)FSSV (SimPO) · AxBench CONCEPT500 (Gemma2-9B)overall steering score: 0.892 -> 0.886No clear effect
use prompt-only intervention (PrOSV) instead of full-sequence (ablation)SV with Lang. objective · AxBench CONCEPT500 (Gemma2-2B)overall steering score: 0.736 -> 0.758Improved
use prompt-only intervention (PrOSV) instead of full-sequence with SimPO (ablation)SV with SimPO objective · AxBench (Qwen2.5-32B)overall steering score: 0.982 -> 1.102Improved
use SimPO objective instead of language modeling (ablation)PrOSV · AxBench (Gemma2-2B)overall steering score: 0.758 -> 0.803Improved
use larger factor initialization size and factor learning rate (ablation)joint SV training · CONCEPT10concept scoreImproved
increase direction initialization size lambda from 1 to 8 (ablation)FSSV · CONCEPT10concept scoreNo clear effect
increase direction initialization size lambda from 1 to 8 (ablation)PrOSV · CONCEPT10concept scoreImproved
intervene on full prompt instead of few tokens (ablation)PrOSV · CONCEPT10 (Gemma2-9B)overall score: 0.92 -> 0.71 (overall); concept 1.14 -> 1.41Hurt
use prefix-suffix p2+s2 instead of p1+s1 or p4+s4 (ablation)PrOSV · CONCEPT10 (Qwen2.5-32B)overall score: 1.10 -> 1.16 (p1+s1 to p2+s2)Improved
use PrOSV instead of FSSV for utilityGemma2-9B steering · tinyGSM8Kaccuracy %: 8.6 -> 68.4Improved

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