Today’s cloud-centric Large Language Model (LLM) inference can raise privacy concerns, as raw user data is sent over public networks. Split LLM inference offers a promising alternative by handling sensitive stages, such as prefill and input/output decoding, locally at the edge while offloading heavy intermediate decoding layers to the cloud. This approach could enhance privacy, yet it introduces a key challenge: how to transfer large intermediate vectors with ultra-low latency and high reliability over dynamic networks. We explore a potential transmission framework combining (i) dynamic congestion control to adapt flow and reduce queueing/loss, and (ii) adaptive FEC and quantization to balance redundancy with compression. Preliminary results indicate that this approach can achieve lower per-token latency and higher throughput than TCP/QUIC, enabling privacy-preserving LLM services that remain highly responsive and aligned with real-time user experience demands.

We present Discoverse, the first unified, modular, open-source 3DGS-based simulation framework for Real2Sim2Real robot learning. It features a holistic Real2Sim pipeline that synthesizes hyper-realistic geometry and appearance of complex real-world scenarios, paving the way for analyzing and bridging the Sim2Real gap. Powered by Gaussian Splatting and MuJoCo, Discoverse enables massively parallel simulation of multiple sensor modalities and accurate physics, with inclusive supports for existing 3D assets, robot models, and ROS plugins, empowering large-scale robot learning and complex robotic benchmarks. Through extensive experiments on imitation learning, Dis coverse demonstrates state-of-the-art zero-shot Sim2Real transfer performance compared to existing simulators. For code and demos: https://air-discoverse.github.io/.

This work reports a novel multi-frame Bundle Adjustment (BA) framework called RKHS-BA. It uses continuous landmark representations that encode RGB-D/LiDAR and semantic observations in a reproducing kernel hilbert space (RKHS). With a correspondence-free pose graph formulation, the proposed system constructs a loss function that achieves more generalized convergence than classical point-wise convergence. We demonstrate its applications in multi-view point cloud registration, sliding-window odometry, and global LiDAR mapping on simulated and real data. It shows highly robust pose estimations in extremely noisy scenes and exhibits strong generalization with various types of semantic inputs.