Cold-Start Personalization via Training-Free Priors from Structured World Models

• URL: http://arxiv.org/abs/2602.15012v1
• Date: Mon, 16 Feb 2026 18:52:13 GMT
• Title: Cold-Start Personalization via Training-Free Priors from Structured World Models
• Abstract summary: Cold-start personalization requires inferring user preferences through interaction when no user-specific historical data is available.<n>We propose decomposing cold-start elicitation into offline structure learning and online Bayesian inference.<n>We achieve 80.8% alignment between generated responses and users' stated preferences versus 68.5% for RL, with 3-5x fewer interactions.
• Score: 90.29121836614081
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Cold-start personalization requires inferring user preferences through interaction when no user-specific historical data is available. The core challenge is a routing problem: each task admits dozens of preference dimensions, yet individual users care about only a few, and which ones matter depends on who is asking. With a limited question budget, asking without structure will miss the dimensions that matter. Reinforcement learning is the natural formulation, but in multi-turn settings its terminal reward fails to exploit the factored, per-criterion structure of preference data, and in practice learned policies collapse to static question sequences that ignore user responses. We propose decomposing cold-start elicitation into offline structure learning and online Bayesian inference. Pep (Preference Elicitation with Priors) learns a structured world model of preference correlations offline from complete profiles, then performs training-free Bayesian inference online to select informative questions and predict complete preference profiles, including dimensions never asked about. The framework is modular across downstream solvers and requires only simple belief models. Across medical, mathematical, social, and commonsense reasoning, Pep achieves 80.8% alignment between generated responses and users' stated preferences versus 68.5% for RL, with 3-5x fewer interactions. When two users give different answers to the same question, Pep changes its follow-up 39-62% of the time versus 0-28% for RL. It does so with ~10K parameters versus 8B for RL, showing that the bottleneck in cold-start elicitation is the capability to exploit the factored structure of preference data.

Related papers

• REAR: Test-time Preference Realignment through Reward Decomposition [68.09214603569744]
  We introduce a novel framework that models the task as a realignment problem, since the base model often fails to align with the stated preference.<n>We derive a REAlignment Reward (REAR) that selectively rescales the proportions of these two reward terms.<n>We show that REAR can be formulated as a linear combination of token-level policy log-probabilities, making it computationally efficient and easy to integrate with various TTS algorithms.
  arXiv  Detail & Related papers  (2026-06-29T14:17:53Z)
• Lightweight Inference-Time Personalization for Frozen Knowledge Graph Embeddings [0.0]
  GatedBias is a lightweight inference-time personalization framework for knowledge graphs.<n>Profile-specific features combine with graph-derived binary gates to produce interpretable, per-entity biases.<n>We evaluate GatedBias on two benchmark datasets.
  arXiv  Detail & Related papers  (2025-12-26T22:30:37Z)
• Robust Preference Alignment via Directional Neighborhood Consensus [13.313830197011983]
  We introduce Robust Preference Selection (RPS), a post-hoc, training-free method by leveraging directional neighborhood consensus.<n>RPS samples multiple responses from a local neighborhood of related preferences to create a superior candidate pool.<n>Our work presents a practical, theoretically-grounded solution for enhancing the reliability of preference-aligned models.
  arXiv  Detail & Related papers  (2025-10-23T12:39:20Z)
• Aligning Deep Implicit Preferences by Learning to Reason Defensively [22.548051297731416]
  We propose Critique-Driven Reasoning Alignment (CDRA) to bridge the preference inference gap.<n>CDRA reframes alignment from a scalar reward-matching task into a structured reasoning process.<n> Experiments demonstrate that CDRA excels at discovering and aligning with users' true preferences while executing robust reasoning.
  arXiv  Detail & Related papers  (2025-10-13T09:26:47Z)
• Personalized Reasoning: Just-In-Time Personalization and Why LLMs Fail At It [81.50711040539566]
  Current large language model (LLM) development treats task-solving and preference alignment as separate challenges.<n>We introduce PREFDISCO, an evaluation methodology that transforms static benchmarks into interactive personalization tasks.<n>Our framework creates scenarios where identical questions require different reasoning chains depending on user context.
  arXiv  Detail & Related papers  (2025-09-30T18:55:28Z)
• Pathways of Thoughts: Multi-Directional Thinking for Long-form Personalized Question Answering [57.12316804290369]
  Personalization is essential for adapting question answering systems to user-specific information needs.<n>We propose Pathways of Thoughts (PoT), an inference-stage method that applies to any large language model (LLM) without requiring task-specific fine-tuning.<n>PoT consistently outperforms competitive baselines, achieving up to a 13.1% relative improvement.
  arXiv  Detail & Related papers  (2025-09-23T14:44:46Z)
• FSPO: Few-Shot Preference Optimization of Synthetic Preference Data in LLMs Elicits Effective Personalization to Real Users [111.56469697145519]
  We propose Few-Shot Preference Optimization, which reframes reward modeling as a meta-learning problem.<n>Under this framework, an LLM learns to quickly adapt to a user via a few labeled preferences from that user, constructing a personalized reward function for them.<n>We generate over 1M synthetic personalized preferences using publicly available LLMs.<n>We evaluate FSPO on personalized open-ended generation for up to 1,500 synthetic users across three domains: movie reviews, pedagogical adaptation based on educational background, and general question answering, along with a controlled human study.
  arXiv  Detail & Related papers  (2025-02-26T17:08:46Z)
• Test-Time Alignment via Hypothesis Reweighting [56.71167047381817]
  Large pretrained models often struggle with underspecified tasks.<n>We propose a novel framework to address the challenge of aligning models to test-time user intent.
  arXiv  Detail & Related papers  (2024-12-11T23:02:26Z)
• Personalized Language Modeling from Personalized Human Feedback [45.16986573937782]
  Personalized large language models (LLMs) are designed to tailor responses to individual user preferences.<n>We propose Personalized-RLHF, an efficient framework that utilizes a lightweight user model to capture individual user preferences.<n>We show that personalized LLMs trained using P-RLHF generate responses that are more closely aligned with individual user preferences.
  arXiv  Detail & Related papers  (2024-02-06T04:18:58Z)
• Linear Alignment: A Closed-form Solution for Aligning Human Preferences without Tuning and Feedback [70.32795295142648]
  Linear alignment is a novel algorithm that aligns language models with human preferences in one single inference step. Experiments on both general and personalized preference datasets demonstrate that linear alignment significantly enhances the performance and efficiency of LLM alignment.
  arXiv  Detail & Related papers  (2024-01-21T10:46:23Z)
• Everyone Deserves A Reward: Learning Customized Human Preferences [25.28261194665836]
  Reward models (RMs) are essential for aligning large language models with human preferences to improve interaction quality. We propose a three-stage customized RM learning scheme, then empirically verify its effectiveness on both general preference datasets and our DSP set. We find several ways to better preserve the general preferring ability while training the customized RMs.
  arXiv  Detail & Related papers  (2023-09-06T16:03:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.