Luyang Zhao

Yitao Jiang, Yaoqing Yang, Luyang Zhao et al.

Long-context Transformer inference increasingly relies on KV-cache compression or quantization. Prior rotation and transform-coding results suggest that the channel basis of each key/value vector affects how faithfully a fixed backend preserves model behavior. We introduce Codec-Gauge, a post-training cache-coordinate layer that learns small orthogonal channel transforms around existing compression and quantization backends. Its frequency-distribution objective combines a token-channel DCT spectral-centroid loss with a smooth rate proxy to concentrate KV energy in low-frequency codec-facing layouts. We evaluate actual compression and decompression using measured bytes and rolling compressed-history scoring. Across six models at $3$, $4$, and $6$ bits/value, learned gauges reduce zfp KL divergence by $44.0\%$ on average relative to raw coordinates and outperform random, Hadamard, DCT, and PCA/KLT controls. The same gauges improve quality preservation for block-uniform and KIVI-style quantization. Experiments on a 27B model and long-context task prompts reproduce the quality trend, while serial storage and timing measurements validate the implemented compressed-cache paths. These results establish cache-coordinate geometry as a practical post-training variable for improving compression fidelity without changing model weights, attention semantics, or backend coding rules.

Weicheng Ma, Luyang Zhao, Chun-Yi She et al.

Recent large language models (LLMs) have demonstrated promising capabilities in modeling real-world knowledge and enhancing knowledge-based generation tasks. In this paper, we further explore the potential of using LLMs to aid in the design of soft modular robots, taking into account both user instructions and physical laws, to reduce the reliance on extensive trial-and-error experiments typically needed to achieve robot designs that meet specific structural or task requirements. Specifically, we formulate the robot design process as a sequence generation task and find that LLMs are able to capture key requirements expressed in natural language and reflect them in the construction sequences of robots. To simplify, rather than conducting real-world experiments to assess design quality, we utilize a simulation tool to provide feedback to the generative model, allowing for iterative improvements without requiring extensive human annotations. Furthermore, we introduce five evaluation metrics to assess the quality of robot designs from multiple angles including task completion and adherence to instructions, supporting an automatic evaluation process. Our model performs well in evaluations for designing soft modular robots with uni- and bi-directional locomotion and stair-descending capabilities, highlighting the potential of using natural language and LLMs for robot design. However, we also observe certain limitations that suggest areas for further improvement.

Luyang Zhao, Yijia Wu, Julien Blanchet et al.

Natural systems integrate the work of many sub-units (cells) toward a large-scale unified goal (morphological and behavioral), which can counteract the effects of unexpected experiences, damage, or simply changes in tasks demands. In this paper, we exploit the opportunities presented by soft, modular, and tensegrity robots to introduce soft lattice modules that parallel the sub-units seen in biological systems. The soft lattice modules are comprised of 3D printed plastic "skeletons", linear contracting shape memory alloy spring actuators, and permanent magnets that enable adhesion between modules. The soft lattice modules are capable of independent locomotion, and can also join with other modules to achieve collective, self-assembled, larger scale tasks such as collective locomotion and moving an object across the surface of the lattice assembly. This work represents a preliminary step toward soft modular systems capable of independent and collective behaviors, and provide a platform for future studies on distributed control.