activity
20242026
collaborators

7 papers

cs.LG2026

RuleSmith: Multi-Agent LLMs for Automated Game Balancing

Ziyao Zeng, Chen Liu, Tianyu Liu +5

Game balancing is a longstanding challenge requiring repeated playtesting, expert intuition, and extensive manual tuning. We introduce RuleSmith, the first framework that achieves…

cs.CV2025

Coffee: Controllable Diffusion Fine-tuning

Ziyao Zeng, Jingcheng Ni, Ruyi Liu +1

Text-to-image diffusion models can generate diverse content with flexible prompts, which makes them well-suited for customization through fine-tuning with a small amount of user-pr…

cs.CV2025

ETA: Energy-based Test-time Adaptation for Depth Completion

Younjoon Chung, Hyoungseob Park, Patrick Rim +7

We propose a method for test-time adaptation of pretrained depth completion models. Depth completion models, trained on some ``source'' data, often predict erroneous outputs when t…

cs.CV2025

ProtoDepth: Unsupervised Continual Depth Completion with Prototypes

Patrick Rim, Hyoungseob Park, S. Gangopadhyay +3

We present ProtoDepth, a novel prototype-based approach for continual learning of unsupervised depth completion, the multimodal 3D reconstruction task of predicting dense depth map…

cs.CV2025

HOMER: Homography-Based Efficient Multi-view 3D Object Removal

Jingcheng Ni, Weiguang Zhao, Daniel Wang +4

3D object removal is an important sub-task in 3D scene editing, with broad applications in scene understanding, augmented reality, and robotics. However, existing methods struggle…

cs.CV2024

RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptions

Ziyao Zeng, Yangchao Wu, Hyoungseob Park +6

We propose a method for metric-scale monocular depth estimation. Inferring depth from a single image is an ill-posed problem due to the loss of scale from perspective projection du…