arXiv:2602.01654v2 Announce Type: replace Abstract: Steering vectors (SVs) offer a lightweight way to control large language models (LLMs) at inference time by shifting hidden activations, providing...
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.
Sparse autoencoders (SAEs) are proposed to extract numerous features from large language model (LLM) representations, yet explaining these features still relies primarily on external observation. This reliance leads to superficial explanations inferred from observed model behavior and computational inefficiency from collecting such behavioral evidence at scale.
arXiv:2608. 02957v1 Announce Type: new Abstract: Steering vectors (SVs) are widely used to influence the expression of concepts (e.
arXiv:2606.18389v2 Announce Type: replace Abstract: Large language models (LLMs) have become an effective tool for synthetic data generation, including for low-resource languages, where generated dat...
arXiv:2506.12576v3 Announce Type: replace Abstract: Sparse autoencoders (SAEs) can enable inference-time topic steering by modifying latent feature activations, but existing steering methods often fa...
arXiv:2608.28806v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) disentangle model activations into interpretable features and are widely used for steering large language models. Most exist...
arXiv:2512. 15134v2 Announce Type: replace-cross Abstract: A goal of interpretability is to recover disentangled representations of latent concepts (features) from the activations of neural networks.
arXiv:2606. 15092v1 Announce Type: new Abstract: Activation steering has emerged as a key methodology for controlling the behavior of large language models (LLMs).
arXiv:2607. 00023v1 Announce Type: cross Abstract: Dense sentence embeddings are fundamental to modern Retrieval-Augmented Generation (RAG) systems but suffer from a lack of interpretability due to feature superposition.
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...
arXiv:2603. 02237v2 Announce Type: replace-cross Abstract: Representation steering offers a lightweight mechanism for controlling the behavior of large language models (LLMs) by intervening on internal activations at inference time.