The paper introduces SYNC, a training‑free inference controller that dynamically selects the optimal layer for injecting persona steering vectors during decoding. SYNC estimates each layer’s sensitivity to every persona using cosine similarity trajectories, allowing the model to adaptively choose the best injection point at each step. Experiments on two models and multiple persona profiles show that SYNC outperforms static multi‑persona steering methods in 21 of 28 configurations, with significant gains in 15 cases.
arXiv:2608. 13482v1 Announce Type: cross Abstract: As language-model-based AI is increasingly deployed in autonomous settings, aligning its goals and values with those of humans becomes critical.
The same Persona behavior can be beneficial in one context but harmful in another, causing static Persona elicitation to perform inconsistently across tasks. We introduce the Persona Selection--Realiz...
arXiv:2608.29798v1 Announce Type: cross Abstract: The same Persona behavior can be beneficial in one context but harmful in another, causing static Persona elicitation to perform inconsistently acros...
The paper introduces Where to Steer (W2S), a framework that adaptively selects which layer of a large language model to intervene on based on the input. It argues that optimal steering layers vary across inputs and demonstrates that W2S outperforms fixed-layer steering methods on several LLMs and alignment tasks, both in‑distribution and out‑of‑distribution. The study highlights input‑dependent layer selection as a complementary design dimension to steering vector extraction.
arXiv:2606. 11201v1 Announce Type: cross Abstract: The wide deployment of LLMs has made model alignment necessary to make newly trained models safely and effectively respond to user instructions.
arXiv:2609.00014v1 Announce Type: cross Abstract: Persona-driven techniques increasingly adapt large language models (LLMs) to diverse contexts. However, existing methods predominantly rely on rigid,...
arXiv:2605. 24395v2 Announce Type: replace Abstract: Alignment plays a fundamental role in many machine learning problems, such as multi-network analysis, multimodal learning, and point cloud registration.
arXiv:2509. 26169v2 Announce Type: replace Abstract: Alignment of large language models remains a central challenge in natural language processing.
arXiv:2608.20344v1 Announce Type: new Abstract: LLM-based "digital twins" aim to simulate how an individual would behavein new environments or respond to novel questions, given some representation of...
arXiv:2407. 21311v2 Announce Type: replace-cross Abstract: Unsupervised domain adaptation (UDA) aims to mitigate domain shift, where the distribution of labeled source data differs from that of unlabeled target data.
arXiv:2607. 10539v1 Announce Type: new Abstract: Existing approaches to infer user traits and generate responses consistent with a persona rely on static prompting.
The paper introduces MATCHA, a method that learns explicit many-to-many layer maps between independently trained language models. MATCHA decomposes the alignment into a layer map—an explicit target‑by‑source matrix—and a shared feature map, allowing joint learning of layer correspondence and feature transformation from prompts. Experiments on 42 model pairs across three families show that MATCHA reconstructs target activations more faithfully and improves retrieval metrics, while enabling transfer of activation‑space interventions between models.