activity
20242026
collaborators

16 papers

cs.CL2026

Style Wins, Substance Loses: A Diagnosis of LLM-as-Judge in Idea Generation

Fengxian Ji, Yuke Li, Jingpu Yang +8

However, whether these judges truly evaluate the scientific substance of ideas or are influenced by superficial stylistic presentation remains an open question. To address this que…

cs.LG2026

Parametric Memory Decoding for Zero-Shot Routing in LoRA-Based External Parametric Memory

Fengxian Ji, Zhuohan Xie, Jingpu Yang +3

With the rise of parametric memory, LoRA-based External Parametric Memory (EPM) has emerged as a modular solution, but existing routing methods often introduce additional training,…

cs.AI2026

Text-Driven 3D Indoor Scene Synthesis in Non-Manhattan Environments

Xianhui Meng, Zirui Song, Yuchen Zhang +10

Large Language Models (LLMs) have demonstrated remarkable capabilities in 3D indoor synthesis for Manhattan environments. However, existing methods often fail to capture plausible…

cs.CV2026

ServImage: An Image Generation and Editing Benchmark from Real-world Commercial Imaging Services

Fengxian Ji, Jingpu Yang, Zirui Song +5

Recent image generation and editing models demonstrate robust adherence to instructions and high visual quality on academic benchmarks. However, their performance on paid, real-wor…

cs.CL2026

The Cylindrical Representation Hypothesis for Language Model Steering

Lang Gao, Jinghui Zhang, Wei Liu +7

Steering is a widely used technique for controlling large language models, yet its effects are often unstable and hard to predict. Existing theoretical accounts are largely based o…

cs.CV2026

TextAlign: Preference Alignment for Text Rendering with Hierarchical Rewards

Mingxuan Cui, Jingpu Yang, Fengxian Ji +6

Faithful text rendering remains a persistent weakness of large text-to-image generative models, as it requires both semantic instruction following and fine-grained glyph-level stru…