12 papers
Distinguishing Imitation Error from Intrinsic Motion Learning Difficulty
Zhaorui Meng, Lu Yin, Xinrui Chen +4
Physics-based motion imitation is central to humanoid control, yet current evaluation metrics (e.g., MPJPE) only quantify imitation outcomes, not their underlying causes. This conf…
Two to Tango: Coupled Task-Reference Selection for Safe LLM Fine-tuning
Xinrui Chen, Jianhao Zhang, Ou Wu +1
Fine-tuning safety aligned large language models (LLMs) on downstream data improves adaptation but may erode learned safety behavior. Existing methods use fixed safety examples, gl…
Is Class Signal Clustered or Routed in Task-Induced Implicit Neural Representation Weight Spaces?
Xinyi Guo, Mingyi He, Haobin Ding +7
Implicit neural representations (INRs) encode images as neural-network weights, making image classification a problem of weight-space classifiability. A natural geometric hypothesi…
Learning to Pose Problems: Reasoning-Driven and Solver-Adaptive Data Synthesis
Yongxian Wei, Yilin Zhao, Zixuan Hu +7
Data synthesis for training large reasoning models offers a scalable alternative to limited, human-curated datasets, enabling the creation of high-quality data. However, existing a…
One Algorithm, Two Goals: Dual Scoring for Parameter and Data Selection in LLM Fine-Tuning
Xinrui Chen, Liu Yang, Ou Wu
In Large Language Model (LLM) fine-tuning, parameter and data selection are common strategies for reducing fine-tuning cost, yet they are typically driven by separate scoring mecha…
Co-Generative De Novo Functional Protein Design
Xinrui Chen, Yizhen Luo, Siqi Fan +1
De novo functional protein design aims to generate protein sequences that realize specified biochemical functions without relying on evolutionary templates, enabling broad applicat…