Quantization, distillation, pruning and serving work aimed at the same accuracy for less memory, latency and money.
arXiv:2506. 22784v2 Announce Type: replace-cross Abstract: Point-pixel registration between LiDAR point clouds and camera images is a fundamental yet challenging task in autonomous driving and robotic perception.
arXiv:2508. 15806v2 Announce Type: replace-cross Abstract: The growing sequence length of large language models poses significant challenges for key-value (KV) caches.
arXiv:2607. 09166v1 Announce Type: new Abstract: Spatial transcriptomics enables profiling of spatial gene expression but is limited by high cost and low throughput, motivating prediction from H&E histopathology images.
arXiv:2607. 09259v1 Announce Type: cross Abstract: Telecom fraud-control studies often stop at detector-level classification, but deployment use requires request-level policy resolution, lifecycle traceability, and auditability.
arXiv:2607. 09202v1 Announce Type: cross Abstract: Continual learning commonly relies on post-hoc mechanisms such as replay, elastic regularization, or distillation.
arXiv:2607. 09063v1 Announce Type: new Abstract: Edge devices are increasingly utilized for deploying deep learning applications on embedded systems.
arXiv:2607. 08978v1 Announce Type: new Abstract: Distributed IoT systems generate multivariate time-series streams for monitoring physical assets, servers, and embedded sensing platforms.
arXiv:2602. 21196v2 Announce Type: replace Abstract: Efficiently processing long sequences with Transformer models usually requires splitting the computations across accelerators via context parallelism.
arXiv:2607. 08786v1 Announce Type: cross Abstract: With the growing deployment of large language models (LLMs), LLM inference cost has become a key challenge.
arXiv:2607. 09385v1 Announce Type: cross Abstract: The growing adoption of large language model-based agents within operating system workflows has increased the importance of energy-efficient inference on laptop-class systems-on-chip (SoCs).
arXiv:2607. 09520v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) are the perceptual backbone of embodied AI, but their energy footprint on edge hardware remains poorly understood.
arXiv:2607. 09052v1 Announce Type: new Abstract: Block sparse attention is a hardware friendly way to alleviate the key-value (KV) cache read bottleneck in large language models (LLMs).
arXiv:2502. 12446v3 Announce Type: replace-cross Abstract: Inference-time intervention (ITI) has emerged as a promising method for steering large language model (LLM) behavior in a particular direction (e.
arXiv:2607. 09287v1 Announce Type: new Abstract: Large language models (LLMs) remain expensive to fine-tune because full-parameter updates require substantial memory, compute, and per-task storage.
arXiv:2607. 08782v1 Announce Type: cross Abstract: Expert parallelism has become the prevailing paradigm to serve Mixture-of-Experts (MoE) models.
arXiv:2607. 09424v1 Announce Type: cross Abstract: We present Soofi S 30B-A3B, a sovereign, open-source Mixture-of-Experts (MoE) hybrid Mamba Transformer foundation model for German and English.
arXiv:2607. 09133v1 Announce Type: cross Abstract: While large-scale text-to-image generative models have achieved unprecedented visual performance, their inherent reliance on multi-step iterative solvers incurs severe inference latency.
arXiv:2509. 22557v5 Announce Type: replace Abstract: Mixed bundle pricing is a classic revenue management problem arising in industries such as e-commerce, tourism, and video games.
arXiv:2607. 09153v1 Announce Type: new Abstract: Process Reward Models (PRMs) have been proven to be highly effective in guiding test-time scaling (TTS) methods, which significantly boost the capabilities of LLM-based multi-agent systems.
arXiv:2607. 08930v1 Announce Type: new Abstract: Efficient serving of diffusion large language models (dLLMs) is hindered by convergence heterogeneity: when batching multiple requests, different sequences converge at different rates, causing faster requests to stall behind slower stragglers and introducing compute bubbles and tail latency.