This paper develops an analytical framework to quantify the spectral-resource and energy costs of semantic communication (SemCom) and derives break-even conditions.

This paper proposes a data-driven method for automatic blind audio equalization using a deep neural network and semantic embeddings.

This paper introduces a structural theorem for the sparsifiability of real-valued codes, which generalizes both combinatorial and continuous notions of sparsification.

This paper improves transformer attention in character-level language modeling by applying FFT-based spectral preprocessing to learned query-key projections.

The paper analyzes token reduction for efficient unified VLM training, finding that while task-specific acceleration saves computation, it destroys the mutual performance gains achieved through joint…

TailLoR is a new parameter-efficient finetuning method that uses the singular bases of pre-trained weights to learn low-rank updates, specifically penalizing updates along dominant directions to impro…

This paper investigates the use of spectral filtering for continuous subgraph matching over dynamic graphs and presents three key findings.

This paper proposes a Row-Movable Active Reconfigurable Intelligent Surface (RM-A-RIS) system for vehicular semantic communication, enhancing spatial diversity and improving Sum-Semantic Spectral Effi…

The paper introduces SB-ECC, a novel score-based decoder that models error correction as continuous-time denoising, achieving state-of-the-art performance across various code families and noise levels…

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…

The paper introduces Fast-TurboQuant, a multiplier-free projection architecture for large language models that uses a structured fast Johnson-Lindenstrauss transform instead of dense matrices, resulti…

The paper proposes SISA (SSM-Informed Softmax Attention), a novel hybrid attention mechanism that integrates state-space model (SSM) importance signals directly into the attention score, achieving sta…

This paper presents a method for compressing matrices using a RePair straight-line program (SLP), allowing matrix-vector products with time and space proportional to the compressed size, and demonstra…

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…

FreqLite introduces an ultra-lightweight, frequency-decomposed linear model that significantly outperforms complex transformers on long-term time-series forecasting while drastically reducing computat…

This paper identifies the root cause of performance degradation in full-duplex Spoken Language Models (SLMs) due to modality interference and proposes Lychee-FD, a framework that decouples conflicting…

This paper introduces a sinusoidal recurrence mechanism for harmonic spectral enrichment in implicit neural representations, validated against various models and tasks.