The paper proposes a theoretically grounded adversarial multi-task learning framework (AMTIDIN) that significantly improves joint interference detection, modulation identification, and interference id…

Physical AI inference (batch-1 decode) is primarily memory-bandwidth-bound, but the observed latency gap between fast and slow GPUs is not solely due to memory bandwidth, as launch-side overheads beco…

This paper characterizes the performance of vision-language model inference on the Qualcomm SM8750 using FastVLM-0.5B as a case study, showing significant speedups and energy savings for different pha…

This paper proposes a unified deep learning framework for joint NBI cancellation and robust soft demodulation using NBI-CNet and LLR-CNet to eliminate error floors and improve performance in OFDM syst…

The paper systematically analyzes the benefits and limits of Attention-FFN Disaggregation (AFD) for Mixture-of-Experts (MoE) LLM serving, demonstrating that AFD is crucial for achieving high throughpu…

This paper proposes a model for cross-band CSI reconstruction in multi-band low-altitude wireless systems using radio-frequency metadata and pilot-guided cross-attention.

This paper addresses the obstacles of using Crossbar-based In-Memory Processing (IMP) accelerators for deep neural networks (DNNs) by reusing bit-shift units for multiplication, applying pruning metho…

This paper proposes a generative AI framework using a cGAN for proactive interference mitigation in ultra-dense indoor networks, achieving significant SINR gain, packet-loss reduction, and CSI oracle…

This paper characterizes the gap between current DNN-based speech enhancement systems and hearing aid constraints, and proposes a lightweight architecture to meet these constraints.

The paper introduces SpikeWFM, a novel hybrid architecture combining spiking neural networks (SNNs) and transformers, which significantly improves the robustness and accuracy of wireless foundation mo…

This paper proposes a physical backdoor attack against deep learning modulation classifiers, utilizing power amplifier non-linear distortions as physical triggers to achieve high attack success rates.

This paper proposes a lightweight, distributed machine learning framework for sub-centimeter indoor localization using D-MIMO in O-RAN architectures, reducing midhaul traffic by 100x while maintaining…

A co-designed hardware-software approach for task-conditional sparsity in multi-task inference models, reducing FLOPs, latency, and energy.

This paper proposes a protocol framework for making the 6G air interface AI-native, focusing on interoperability and preserving implementation freedom.