collaborators

8 papers

cs.CV2026

DualMem: Bypassing the Objectness Bottleneck for Calibrated Unknown-Stream Filtering in Open-World Object Detection

Yingjun Xiao, Xi Chen, Gang Fang +1

Open-world object detection (OWOD) requires detectors to localize known classes while identifying unknown objects for future incremental learning. We find that the unknown predicti…

cs.LG2026

Attention Sinks and Outliers in Attention Residuals

Haozheng Luo, Haoran Dai, Shaoyang Zhang +10

We propose OASIS, an outlier- and sink-aware technique built on inter-layer null signaling. As AttnResidual architectures introduce an additional depth-wise normalization channel,…

cs.CV2026

Evidence-Guided Unknown Rejection for High-Confidence Near-Known Unknowns

Xi Chen, Yingjun Xiao, Gang Fang

Open-set recognition systems face a neglected failure mode: high-confidence near-known unknowns, which lie outside the known label set but are close enough to known classes that a…

cs.LG2026

Optimal low-rank stochastic gradient estimation for LLM training

Zehao Li, Tao Ren, Zishi Zhang +2

Large language model (LLM) training is often bottlenecked by memory constraints and stochastic gradient noise in extremely high-dimensional parameter spaces. Motivated by empirical…

cs.LG2026

Astro: Activation-guided Structured Regularization for Outlier-Robust LLM Post-Training Quantization

Xi Chen, Ming Li, Junxi Li +5

Weight-only post-training quantization (PTQ) is crucial for efficient Large Language Model (LLM) deployment but suffers from accuracy degradation caused by weight and activation ou…

cs.LG2026

Hyperparameter Transfer Laws for Non-Recurrent Multi-Path Neural Networks

Shenxi Wu, Haosong Zhang, Xingjian Ma +4

Deeper modern architectures are costly to train, making hyperparameter transfer preferable to expensive repeated tuning. Maximal Update Parametrization (P) helps explain why ma…