5 papers
LoCA: Forward-Only LLM Tuning after One-Shot Calibration with Local Credit Assignment
Linhan Xia, Rui Liu, Zhaofeng Zhang +3
Parameter-efficient post-training reduces the number of trainable parameters, but still requires repeated end-to-end backpropagation through the frozen backbone. Every adaptation s…
Evolving in the Agent Jungle via History-Informed Opponent Awareness
Zhaofeng Zhang, Linhan Xia, Rui Liu +3
Learning to adapt strategies through interaction is a key step toward more general and autonomous LLM agents. Existing approaches typically achieve behavioral adaptation by revisin…
FRAM: Frobenius-Regularized Assignment Matching with Mixed-Precision Computing
Binrui Shen, Yuan Liang, Shengxin Zhu
Graph matching, typically formulated as a Quadratic Assignment Problem (QAP), seeks to establish node correspondences between two graphs. To address the NP-hardness of QAP, some ex…
Adaptive Softassign via Hadamard-Equipped Sinkhorn
Binrui Shen, Qiang Niu, Shengxin Zhu
Softassign is a pivotal method in graph matching and other learning tasks. Many softassign-based algorithms exhibit performance sensitivity to a parameter in the softassign. Howeve…
CSGO: Constrained-Softassign Gradient Optimization For Large Graph Matching
Binrui Shen, Qiang Niu, Shengxin Zhu
Graph matching aims to find correspondences between two graphs. This paper integrates several well-known graph matching algorithms into a framework: the constrained gradient method…