…View arXiv page View PDF Add to collection Community A differentiable optimal transport framework for dense-to-MoE model conversion; retaining 90% of dense performance at 50% active parameters. Get this paper…
…Although the drafter network is small in modern architectures, its LM-head still performs projection to a large vocabulary, becoming one of the major computational bottlenecks. In prior work this issue has…
…In this work, we empirically investigate nonuniform capacity allocation across network depth by proposing a times-shaped >
…Given a query, an LLM-based classifier identifies the relational intent, and a routing network dynamically modulates the corresponding dimensions of the edge embedding . Traversal scores are computed via a learned combination…
…View arXiv page View PDF Project page Add to collection Community This paper studies why extremely large activations suddenly appear inside large language models. We identify a specific layer, called the Massive…
…We have implemented bilinear interpolation , movable kernels , and tiny neural networks as spatially varying functions. SVGS employs 2D Gaussian surfels as primitives, which significantly enhances novel-view synthesis while maintaining high-quality…
…Architecturally, NTM combines shallow invertible blocks within each step with a deep parallel predictor across the trajectory, forming an end-to-end network trainable from scratch or initializable from pretrained flow-matching…
…GitHub 5 Add to collection Community Lane-level maps are critical infrastructure for autonomous driving and lane-level navigation, yet constructing and maintaining standardized lane networks for hundreds of cities remains highly…
…Experiments on three LIBERO benchmark suites demonstrate that RL4IL substantially outperforms state-of-the-art imitation learning methods under sensor dropout conditions, while requiring no policy network training. The code can be…
…Motivated by the spectral bias of neural networks and the principles of curriculum learning , our method progressively refines discretization per feature upon plateau detection and selects representation-aware splits to jointly improve…