AI Paper Atlas

Research map / Large language models / Alignment and preferences

Simulated Rewards, Skewed Strategies: Tracing the Acquired Preference Bias in LLM-Based Dialogue Planners

Heyan Huang, Yizhe Yang, Huashan Sun, Jiawei Li, Yang Gao

AAAI 2026 · Large language models › Alignment and preferences · approach: Reward model design and preference models · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
RL-trained planners (PPDPP, TRIP) vs prompt-only Standarddialogue planners · Persuasion, GPT-4.1preference bias B: 1.02 -> 2.92 (PPDPP)Hurt
RL-trained PPDPP vs Standard promptingdialogue planner · Persuasion, GPT-4.1strategy entropy: 2.08 -> 0.26Hurt
search-based GDPZero/DPDP vs Standarddialogue planners · Emotional Support, Qwen-32Bpreference bias B: 0.73 -> 0.58 (GDPZero)Improved
GDPZero/DPDP vs Standard in negotiation and persuasiondialogue planners · Price Negotiation, GPT-4.1preference bias B: 1.22 -> 1.52 (GDPZero)Mixed
change the user-simulator LLM (ablation)dialogue planners · role decomposition, negotiationconvergence score: 0.6-0.8 -> 0.2-0.4Hurt
add persona diversification to user simulator (ablation)GDPZero · Emotional Support, GPT-4.1preference bias B: 1.20 -> 2.17Hurt
add persona diversification to user simulator (ablation)PPDPP · Price Negotiation, GPT-4.1preference bias B: 1.64 -> 2.20Hurt
add persona diversification to user simulator (ablation)PPDPP · Persuasion, GPT-4.1preference bias B: 2.92 -> 2.97No clear effect
add inappropriate strategies to the action spacePPDPP and one other planner · negotiation, support, persuasioninappropriate-strategy adoptionHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy