<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>Publications on Kaijie Xu</title><link>https://nortrom1213.github.io/publications/</link><description>Recent content in Publications on Kaijie Xu</description><generator>Hugo -- 0.147.2</generator><language>en</language><lastBuildDate>Wed, 17 Jun 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://nortrom1213.github.io/publications/index.xml" rel="self" type="application/rss+xml"/><item><title>Action Selection in Multiplayer Hidden-Information Games with Expected Target Entropy</title><link>https://nortrom1213.github.io/publications/2026-cog-expected-target-entropy/</link><pubDate>Wed, 17 Jun 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-cog-expected-target-entropy/</guid><description>A reusable belief-state adapter and one-step expected-target-entropy policy evaluate information-seeking action selection in multiplayer hidden-information games.</description></item><item><title>Boss Fights as Spatio-Temporal Level Generation</title><link>https://nortrom1213.github.io/publications/2026-cog-boss-fight-generation/</link><pubDate>Wed, 17 Jun 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-cog-boss-fight-generation/</guid><description>A constrained time-expanded representation coordinates spatial navigation and evolving hazards for auditable 2D boss-fight generation.</description></item><item><title>Efficient Multi-Guard Patrol through Extended Composite Potential Fields</title><link>https://nortrom1213.github.io/publications/2026-cog-efficient-multi-guard-patrol/</link><pubDate>Wed, 17 Jun 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-cog-efficient-multi-guard-patrol/</guid><description>Three modular CPF extensions improve real-time efficiency, spatial coordination, and area coverage for multi-guard stealth patrols.</description></item><item><title>Scene-Modular LLM-Driven Procedural Level Generation for Anomaly-Detection Games</title><link>https://nortrom1213.github.io/publications/2026-cog-scene-modular-anomaly-generation/</link><pubDate>Wed, 17 Jun 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-cog-scene-modular-anomaly-generation/</guid><description>A scene-modular framework uses local open-source LLMs to generate structured, difficulty-controllable anomalies for fixed 3D environments.</description></item><item><title>Generate Diverse Skills with Large Language Models</title><link>https://nortrom1213.github.io/publications/2026-fdg-generate-diverse-skills-llms/</link><pubDate>Sat, 09 May 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-fdg-generate-diverse-skills-llms/</guid><description>Neural Spore Genesis evaluates local open-weight LLMs for generating coherent and diverse auto-battler combat skills from spatial elemental layouts.</description></item><item><title>Path2Patrol: Path-Conditioned Guard Generation in Stealth Games through MAP-Elites</title><link>https://nortrom1213.github.io/publications/2026-cog-path2patrol/</link><pubDate>Sat, 09 May 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-cog-path2patrol/</guid><description>Path2Patrol uses MAP-Elites to generate guard patrols conditioned on a designer-specified player solution path for stealth games.</description></item><item><title>Deconstructing Open-World Game Mission Design Formula: A Thematic Analysis Using an Action-Block Framework</title><link>https://nortrom1213.github.io/publications/2026-chi-mission-design/</link><pubDate>Thu, 19 Mar 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-chi-mission-design/</guid><description>A six-dimensional Mission Action Quality Vector (MAQV) framework and action block grammar for systematic analysis of open-world mission design at scale.</description></item><item><title>(Perlin) Noise as AI coordinator</title><link>https://nortrom1213.github.io/publications/2026-fdg-perlin-noise-ai-coordinator/</link><pubDate>Sat, 21 Feb 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-fdg-perlin-noise-ai-coordinator/</guid><description>A general framework that uses continuous noise fields (e.g., Perlin noise) as a scalable coordinator for large-scale nonplayer agent control.</description></item><item><title>High Dimensional Procedural Content Generation</title><link>https://nortrom1213.github.io/publications/2026-fdg-hdpcg/</link><pubDate>Sat, 21 Feb 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-fdg-hdpcg/</guid><description>We introduce High-Dimensional PCG (HDPCG), treating gameplay-relevant dimensions beyond geometry as first-class coordinates for controllable, verifiable generation.</description></item><item><title>How Far Can We Go with Pixels Alone? A Pilot Study on Screen-Only Navigation in Commercial 3D ARPGs</title><link>https://nortrom1213.github.io/publications/2026-fdg-pixels-alone-navigation/</link><pubDate>Sat, 21 Feb 2026 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-fdg-pixels-alone-navigation/</guid><description>A screen-only exploration and navigation agent driven by visual affordances for Dark Souls-style linear 3D ARPG levels.</description></item><item><title>CSP4SDG: Constraint and Information-Theory Based Role Identification in Social Deduction Games with LLM-Enhanced Inference</title><link>https://nortrom1213.github.io/publications/2026-aaai-csp4sdg/</link><pubDate>Wed, 12 Nov 2025 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2026-aaai-csp4sdg/</guid><description>A probabilistic, constraint-satisfaction framework with information-theoretic weighting for interpretable role inference in social deduction games.</description></item><item><title>A Database-Driven Framework for 3D Level Generation with LLMs</title><link>https://nortrom1213.github.io/publications/2025-11-03/</link><pubDate>Sat, 01 Nov 2025 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2025-11-03/</guid><description>A modular, LLM-assisted, database-centric pipeline that generates coherent multi-floor 3D levels through constraint-based optimization.</description></item><item><title>Adaptive Visual Navigation Assistant in 3D RPGs</title><link>https://nortrom1213.github.io/publications/2025-11-02/</link><pubDate>Sat, 01 Nov 2025 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2025-11-02/</guid><description>A two-stage deep-learning pipeline that detects and ranks map transition points, enabling AI-driven navigation aids in complex 3D RPG environments.</description></item><item><title>Generic Guard AI in Stealth Games with Composite Potential Fields</title><link>https://nortrom1213.github.io/publications/2025-11-01-guard-ai/</link><pubDate>Sat, 01 Nov 2025 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2025-11-01-guard-ai/</guid><description>A generic, training-free composite-potential-field framework that yields responsive and believable guard AI for stealth games.</description></item><item><title>Action Window Planning for Stealth Missions</title><link>https://nortrom1213.github.io/publications/2025-08-p2/</link><pubDate>Fri, 01 Aug 2025 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2025-08-p2/</guid><description>Two novel methods for generating spatiotemporal action windows in stealth mission planning via evolutionary optimization and safeblock search.</description></item><item><title>Constraint Propagation for Reasoning in Single-player Deduction Games</title><link>https://nortrom1213.github.io/publications/2025-08-p3/</link><pubDate>Fri, 01 Aug 2025 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2025-08-p3/</guid><description>A CSP-based variant of Information-Set Entropy Search using AC-3 for efficient inference in single-player deduction games.</description></item><item><title>Quantitative Analysis of Visual Guidance in Level Transitions Using Multimodal Visual Metrics</title><link>https://nortrom1213.github.io/publications/2025-08-p1/</link><pubDate>Fri, 01 Aug 2025 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2025-08-p1/</guid><description>A bimodal, data-driven framework quantifying visual cues in 3D RPG level transitions.</description></item><item><title>Constraint Is All You Need: Optimization-Based 3D Level Generation with LLMs</title><link>https://nortrom1213.github.io/publications/2025-04-1/</link><pubDate>Tue, 01 Apr 2025 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2025-04-1/</guid><description>An LLM-driven pipeline that parses narrative descriptions into a structured GLDL and solves level generation as a facility layout optimization problem.</description></item><item><title>FedFDD: Federated Learning with Frequency Domain Decomposition for Low-Dose CT Denoising</title><link>https://nortrom1213.github.io/publications/2024/</link><pubDate>Mon, 01 Jul 2024 00:00:00 +0000</pubDate><guid>https://nortrom1213.github.io/publications/2024/</guid><description>A personalized federated learning approach that decomposes LDCT images into frequency components for robust denoising across heterogeneous clients.</description></item></channel></rss>