arXiv Machine Learning October 15, 2026 By Tora Bodin, Eric Aubinais, Ilaria Torre, Chih-Hong Cheng

Genesis Steering: Better Activation Steering Through Training Data

Read the original on arXiv Machine Learning →

The paper introduces genesis steering, a method that derives a steering vector from contrastive examples in a model’s own fine‑tuning data, and proposes using membership inference to build proxy datasets when such data are unavailable. Two theoretical results are presented on optimal displacement to the decision boundary and reduced proxy‑estimation bias. Experiments on T"ulu‑3‑8B and Llama‑3.1‑8B‑Instruct show that safety data from T"ulu‑3‑8B’s training mixture produce stronger steering than external datasets, and that membership‑based proxy construction can achieve competitive performance.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Sep 3

IDEEA: training-free Input-Dependent stEEring via Activation cluster matching

IDEA is a training‑free, input‑dependent steering method for large language models that matches activations to cluster‑specific directions aligned with a target concept. It clusters positive and negative activation supports per attention head, solves an optimal‑matching problem to create a pool of cluster‑conditional directions, and selects the best match for each input at inference time. This approach preserves the input’s original representation while improving the truth × info rate on TruthfulQA by an average of 9.9% (up to 23.5%) over input‑independent baselines.

By Zheng Wang, Muchen Li, Renjie Liao, Yan Leng

llms reinforcement-learning fine-tuning safety

Hugging Face Trending Papers

Sep 2

IDEEA: training-free Input-Dependent stEEring via Activation cluster matching

IDEA: training-free Input-Dependent stEEring via Activation cluster matching (IDEEA) is a method that steers large language models by injecting bias into selected activations at inference time, without requiring weight updates. Unlike existing training-free steering approaches that use a single, input-independent direction, IDEEA clusters positive and negative activation supports per attention head and solves an optimal-matching problem to create a set of cluster-conditional directions. At inference, IDEEA selects the direction that best matches the input’s activation, aligning the model toward a target concept while preserving the input’s original representation, and achieves a 9.9% average improvement in truth × info rate on TruthfulQA compared to the best input-independent baseline.

llms reinforcement-learning fine-tuning safety

arXiv AI

Jul 13

GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs

arXiv:2507. 18043v2 Announce Type: replace-cross Abstract: Inference-time steering methods offer a lightweight alternative to fine-tuning large language models (LLMs) and vision-language models (VLMs) by modifying internal activations at test time without updating model weights.

By Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal

llms fine-tuning multimodal safety

arXiv Machine Learning

Jul 9

Towards Understanding Steering Strength

arXiv:2602. 02712v2 Announce Type: replace Abstract: A popular approach to post-training control of large language models (LLMs) is the steering of intermediate latent representations.

By Magamed Taimeskhanov, Samuel Vaiter, Damien Garreau

llms

arXiv AI

Oct 15

Contrastive Covariance Subspaces for Activation Steering

arXiv:2610.14025v1 Announce Type: cross Abstract: Activation steering controls large language model behavior by adding a direction to hidden activations during inference. For steering to work well, t...

By Alexandros Xenos, Manos Plitsis, Niki Foteinopoulou, Yannis Panagakis, Ioannis Patras

llms benchmarks

arXiv AI

Sep 15

Signatures of Steerability in Activation Space of Language Models

arXiv:2609.14151v1 Announce Type: cross Abstract: Steering language models using a set of contrastive representations has been a canonical and computationally efficient method for controlling model b...

By Prajjwal Bhattarai, Tuka Alhanai

llms safety

One email a morning, machine-written

Related stories

IDEEA: training-free Input-Dependent stEEring via Activation cluster matching

IDEEA: training-free Input-Dependent stEEring via Activation cluster matching

GrAInS: Gradient-based Attribution for Inference-Time Steering of LLMs and VLMs

Towards Understanding Steering Strength

Contrastive Covariance Subspaces for Activation Steering

Signatures of Steerability in Activation Space of Language Models