AI Paper Atlas

Research map / Large language models / Alignment and preferences

VORTEX: Aligning Task Utility and Human Preferences Through LLM-Guided Reward Shaping

Guojun Xiong, Milind Tambe

AAAI 2026 · Large language models › Alignment and preferences · approach: Active and token-level preference learning · 2 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
add a human preference via LLM reward shapingBase Reward policy · Public health simulationtotal utility: 432 (median, base reward) then lower under any preferenceHurt
apply Favor LI reward shapingBase Reward policy · Public health simulationpull proportion of low-income: 0.22 -> 0.75Improved
apply one reflection step with verbal feedbackVORTEX · Public health illustrative trajectoryutility: 82.4 -> 80.7 (-1.7)Hurt
apply one reflection step with verbal feedbackVORTEX · Public health illustrative trajectorycoverage: 22.0% -> 27.5% (+5.5%)Improved
replace DLM with VORTEX reflective reward shapingDLM baseline · Public health, Favor LI and Favor LEtotal utility stabilityImproved
replace DLM with VORTEX reflective reward shapingDLM baseline · Public health, low-income coveragecoverage ratioImproved

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