Training large language models faces frequent interruptions due to various faults, demanding robust fault-tolerance. Existing backup-free methods, such as redundant computation, dynamic parallelism, and data rerouting, each incur performance penalties, whether from ongoing overhead, lengthy reconfigurations, or post-recovery inefficiencies. We propose Chameleon, an adaptive fault-tolerant system that intelligently selects optimal recovery strategies when a failure occurs. Chameleon achieves this through a unified performance model, expedient execution plan search, accurate performance estimation, and efficient communication optimizations. Experiments on a 32-card cluster show that Chameleon maintains a performance gap of within 11.00% between post-recovery and failure-free training, while preserving model convergence and efficient memory usage. Compared to state-of-the-art methods, Chameleon achieves up to 1.229x and 1.355x higher average throughput than Oobleck and Recycle, respectively.

Jitter is a critical factor affecting the geometric accuracy of Earth observation satellites. After a satellite is launched, it is essential to detect and mitigate platform jitter to minimize its impact. The Ziyuan-3 (ZY3) series, China’s first civilian stereo mapping satellite constellation, meets the 1:50000 scale mapping accuracy requirement without ground control points. Detecting and suppressing jitter during the satellite’s orbit is a key technology for ensuring mapping accuracy. This letter investigates the causes of platform jitter across three satellites in the ZY3 series by analyzing gyroscope data using fast Fourier transform (FFT) at different stages of the satellite lifecycle. It compares the frequency and amplitude of jitter across different platforms and time periods. The results show that ZY3-01 and ZY3-02 exhibit consistent jitter patterns. They have significant amplitude in the 0.6 Hz region early in orbit and less later on, correlating with the satellites’ lifespans. In contrast, the 0.2 Hz region remains relatively stable. The ZY3-03 satellite, equipped with a highly stable platform, shows no significant jitter, evidencing the effectiveness of such platforms in jitter suppression.

We present an adaptive multi-feature fusion visual object tracking algorithm based on Siamese neural network with cross-attention mechanism, SiamAtten for short, which can effectively deal with large appearance changes, complex back-ground and interference. The proposed network consists of two parts. One is the full convolution feature extraction network based on cross-attention mechanism and the other is the region proposal generation network based on adaptive multi-feature fusion. The cross-attention mechanism is used in the first part to improve the response ability of feature extraction of the target. The adaptive feature fusion method is used in the second part to infer the target location in a step-by-step process and to get the robust region proposal by regression. Meanwhile, network parameters are reduced by using the depth-wise separable convolution, and the cross-attention mechanism is proposed in this paper can effectively enhance the target identification ability and elevate the robustness. Extensive experiments are carried out on three benchmark datasets and much advanced tracking results are obtained.