4 papers
TEPA: Revoking Stale Memories for Conflict-Robust Language Agents
Yan Zhou, Yue Ouyang, Kaiyang Zheng +1
Long-term memory enables language agents to reuse past facts, preferences, and task experience. Persistence also creates a central falsifiability problem: when the world changes, s…
ExFusion: Efficient Transformer Training via Multi-Experts Fusion
Jiacheng Ruan, Daize Dong, Xiaoye Qu +5
Mixture-of-Experts (MoE) models substantially improve performance by increasing the capacity of dense architectures. However, directly training MoE models requires considerable com…
Colo-ReID: Discriminative Representation Embedding with Meta-learning for Colonoscopic Polyp Re-Identification
Suncheng Xiang, Chengfeng Zhou, Zhengjie Zhang +2
Colonoscopic Polyp Re-Identification aims to match the same polyp from a large gallery with images from different views taken using different cameras and plays an important role in…
Understanding Robustness of Parameter-Efficient Tuning for Image Classification
Jiacheng Ruan, Xian Gao, Suncheng Xiang +3
Parameter-efficient tuning (PET) techniques calibrate the model's predictions on downstream tasks by freezing the pre-trained models and introducing a small number of learnable par…