papers

Publications (9)

cs.CL2025

Magnifier Prompt: Tackling Multimodal Hallucination via Extremely Simple Instructions

Yuhan Fu, Ruobing Xie, Jiazhen Liu +4

Hallucinations in multimodal large language models (MLLMs) hinder their practical applications. To address this, we propose a Magnifier Prompt (MagPrompt), a simple yet effective m…

cs.CL2024

Mitigating Hallucination in Multimodal Large Language Model via Hallucination-targeted Direct Preference Optimization

Yuhan Fu, Ruobing Xie, Xingwu Sun +2

Multimodal Large Language Models (MLLMs) are known to hallucinate, which limits their practical applications. Recent works have attempted to apply Direct Preference Optimization (D…

cs.CV2025

PhD: A ChatGPT-Prompted Visual hallucination Evaluation Dataset

Jiazhen Liu, Yuhan Fu, Ruobing Xie +5

Multimodal Large Language Models (MLLMs) hallucinate, resulting in an emerging topic of visual hallucination evaluation (VHE). This paper contributes a ChatGPT-Prompted visual hall…

hep-th2026

Notes on Diagrammatic Coaction for Cosmological Wavefunction Coefficients: A Two-Site Prelude

Yuhan Fu, Jiahao Liu

We study the coaction of cosmological wavefunction coefficients of conformally coupled scalars in FRW background of a two-site example, which turns out to have an elegant diagramma…

cs.CL2026

PILA: Plug-and-Play Insertion for LLM-native Advertising

Zhaowei Zhang, Yuhan Fu, Yihang Zhang +6

The paper introduces PILA, a plug‑and‑play sidecar that rewrites LLM responses to insert sponsored content, allowing ad placement without modifying the underlying language model.

#advertising#large language models#response rewriting#plug-and-play
cs.LG2025

Towards a Comprehensive Scaling Law of Mixture-of-Experts

Guoliang Zhao, Yuhan Fu, Shuaipeng Li +10

Mixture-of-Experts (MoE) models have become the consensus approach for enabling parameter-efficient scaling and cost-effective deployment in large language models. However, existin…

cs.AI2026

Evaluating and Pricing Advertisements in AI-Generated Responses

John L. Turner-Smith, Zimeng Huang, Yuhan Fu +2

The paper introduces a psychologically grounded agent simulation to create supervision for predicting click‑through intent of ads embedded in LLM‑generated responses, builds a ligh…

#advertising#large language models#click-through prediction#pricing mechanisms
astro-ph.HE2022

The disk wind in GRS 1915+105 as seen by Insight-HXMT

Honghui Liu, Yuhan Fu, Cosimo Bambi +7

We analyze three observations of GRS 1915+105 in 2017 by Insight-HXMT when the source was in a spectrally soft state. We find strong absorption lines from highly ionized iron, whic…

cs.CV2025

Multi-Object Sketch Animation by Scene Decomposition and Motion Planning

Jingyu Liu, Zijie Xin, Yuhan Fu +3

Sketch animation, which brings static sketches to life by generating dynamic video sequences, has found widespread applications in GIF design, cartoon production, and daily enterta…