The study investigates how large language models (LLMs) encode persona-dependent preferences by training linear probes on residual-stream activations of Gemma‑3‑27B and Qwen‑3.5‑122B. It identifies a genuine preference vector that tracks the model’s task choices across various prompts and shows that steering along this vector can causally control pairwise choices. The research also finds that some preference information transfers between different personas, including a case where an evil persona’s preferences anti‑correlate with those of a helpful assistant.
arXiv:2609.23935v1 Announce Type: new Abstract: Post-training turns a general next-token predictor into a chat model with a persistent assistant persona. If that persona is a character the model play...
arXiv:2606. 22974v2 Announce Type: replace Abstract: Recent work on preference elicitation in large language models (LLMs) has demonstrated that, when given a series of choices between two outcomes, LLMs reveal a coherent, model-specific utility structure.
arXiv:2509. 03647v2 Announce Type: replace-cross Abstract: Large language models (LLMs) increasingly serve as automated evaluators, yet they suffer from "self-preference bias": a tendency to favor their own outputs over those of other models.
arXiv:2608.30902v1 Announce Type: new Abstract: Adapting large language models to user-specific preferences is often constrained by the cost of human annotation, making preference optimisation imprac...
The paper introduces the Atomic User Model (AUM), a structured representation of a user’s personality that separates a stable identity nucleus from four interpretable shells—psychological, cognitive & experiential, behavioural, and social—along with cross-shell entries for conflict and authenticity. It proposes using AUM as a retrieval index rather than a prompt prefix, enabling a task‑specific, budgeted retrieval of relevant fields at generation time. Experiments with simulated participants show that retrieving eight AUM fields improves style fidelity, preference accuracy, and user voice identification compared to flat preference notes, especially benefiting users whose default assistant performs poorly.
arXiv:2607. 13162v1 Announce Type: cross Abstract: What a language model will and will not do is largely set during post-training, but which behaviors it expresses, hides, or resists is not revealed by prompting alone.
arXiv:2606. 00334v1 Announce Type: cross Abstract: Various language domains have undergone remarkable changes in recent years; these shifts are largely attributed to the advent of Large Language Models and their misalignment with natural language usage.
arXiv:2609.39882v1 Announce Type: new Abstract: Pre-training equips large language models (LLMs) with a broad repertoire of behavioral patterns associated with roles, styles, values, and goals. Post-...
arXiv:2601. 02896v3 Announce Type: replace Abstract: Controlling emergent behavioral personas (e.
arXiv:2605. 09159v2 Announce Type: replace Abstract: Recent work shows that large language models (LLMs) encode behavioral traits ("personas") as linear directions in activation space, often called "persona vectors".
arXiv:2608.31079v1 Announce Type: new Abstract: Sycophantic agreement refers to a behavior in which language models excessively affirm the user, often at the cost of factual accuracy. Although sycoph...
arXiv:2608. 05624v1 Announce Type: new Abstract: Sycophantic responses are becoming pervasive in large language models (LLMs), and prior work has pointed out that some of them could be harmful.