Clay weekly context brief for the Systems category (ISO week 2026-W45).

Clay tracks publications from the Systems feed list. Below are recent items from this category, each with its source and a short description of what the publication covers when one is available in the source feed.

Recent publications:
1. Medical Image Analysis with Machine Learning (MEDiml): Strategies for Simpler Radiomics
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.37715
   Radiomics extracts quantitative features describing tumor shape, intensity, and texture from medical images to build models that predict clinical outcomes.
2. StreamSE: Neural Audio Codec-based Streaming Speech Enhancement for On-device Inference
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2610.37024
   Recent generative speech enhancement (SE) models achieve high perceptual quality, but their non-causal processing and expensive computation make them difficult to use for streaming, on-device inference.
3. Multichannel Sampling and Reconstruction in Multidimensional Shift-Invariant Spaces
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.36142
   We develop a multichannel sampling and reconstruction framework for multidimensional principal shift-invariant spaces based on a finite-dimensional polyphase representation.
4. EUReCA: End-to-End Unified CBCT Reconstruction Across Clinical Acquisitions
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.37740
   Cone-beam computed tomography (CBCT) is widely used for image-guided radiotherapy, but reducing the number of projections to lower dose and scan time makes reconstruction increasingly ill-posed.
5. Developing a Speech Language Model for Basque
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2610.36911
   Although the recent development of LLMs has paved the way for multi-modal interaction, minority languages face difficulties keeping pace.
6. MtReNIRS: A Self-Supervised Foundation Model for Transferable Neurovascular Representations
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.36159
   Functional near-infrared spectroscopy (fNIRS) offers a promising approach for monitoring the hemodynamic correlates of neural activity in naturalistic settings.
7. Flow Matching for Fluctuation-Based Fluorescence Super-Resolution
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.37802
   Fluctuation-based super-resolution techniques are an appealing alternative to single-molecule and structured-illumination microscopy, as they adapt to standard acquisition settings and exploit the independent stochastic fluctuations of fluorophores across a sequence of frames.
8. Integrated Spoofing-Robust Automatic Speaker Verification via a Three-Class Formulation and LLR
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2603.13780
   Spoofing-robust automatic speaker verification (SASV) aims to integrate automatic speaker verification (ASV) and countermeasure (CM).
9. Structured Wavenumber-Domain Channel Estimation for LoS XL-MIMO
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.36609
   Near-field line-of-sight (LoS) channels in extremely large-scale multiple-input multiple-output (XL-MIMO) systems exhibit distinctive structural properties in the wavenumber domain.
10. Looping Kooshball: Improved k-space coverage for silent fMRI using 3D coherence trajectories
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.38072
   Functional MRI (fMRI) is widely used to characterize brain activity and connectivity, but conventional gradient-echo echo-planar imaging produces high acoustic sound-pressure levels by operating gradient hardware near its limits.
11. Sonifying I2S Transport Signals to Detect Transmission Faults
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2608.16498
   This paper outlines a sonification design to support fault detection in the transmission of I2S transport signals.
12. Soft Temporal Alignment Distillation for Camera-to-mmWave Radar Human Activity Recognition
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.36637
   Millimeter-wave (mmWave) radar provides privacy?preserving human activity recognition (HAR), but sparse point clouds limit its representation of body structure and motion.
13. From the Superderivative to a Relative-Gradient Edge Operator
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.38530
   First-derivative edge detectors such as Sobel, Prewitt, Roberts and Scharr measure absolute grey-level differences.
14. AzeroS: Extending LLM to Speech with Self-Generated Instruction-Free Tuning
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2601.06086
   Extending large language models (LLMs) to the speech domain has recently gained significant attention.
15. Static Multipath Suppression in Power Delay Profile of Time-Varying Channel at 28 GHz
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.36792
   This paper investigates static multipath suppression in a time-varying radio channel at 28 GHz using wideband measurements conducted in a dynamic road environment.
16. Guidance-Coupled Nullspace Drifting for One-Step Medical Image Reconstruction
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.38953
   Generative reconstruction can recover high-quality images from undersampled acquisitions, but state-of-the-art generative priors commonly rely on iterative sampling.
17. A Benchmark for Controllable Speaking-Style Captioning in Audio-Language Models
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2610.33036
   Speaking-style captioning (SSC) aims to generate natural language descriptions of how speech is delivered, capturing paralinguistic attributes such as vocal timbre, prosody, and expressivity.
18. Spatial MIMO Channel Rendering: A 3D Gaussian Splatting Approach
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.36880
   The achievable gains of multiple-input multiple-output (MIMO) systems depend on channel state information (CSI), whose acquisition consumes pilot and feedback resources for every user in every coherence interval.
19. Blind Video Denoising for Vocal Tract rtMRI using a Self-Supervised Ensemble
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.39972
   Real-time magnetic resonance imaging of the vocal tract makes speech articulation directly observable, but the trade-off between temporal resolution and signal quality introduces substantial noise and obscures fine articulatory detail.
20. Encode Once, Decode Never: Reusing Audio LM Internals for Efficient Temporal Localization
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2602.10230
   Audio language models process input audio into rich frame-level representations, but the standard approach to temporal localization generates timestamps as sequences of text tokens, which discards the frame-level representations in favor of autoregressive decoding.
21. RAG-CSI: Global Local Retrieval via Joint State and Dynamics Matching for CSI Prediction
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.36881
   Channel state information (CSI) prediction forecasts future channel states from historical observations, supporting transceiver design and resource allocation with reduced measurement overhead.
22. Effects of Deep Learning Architecture and Pretraining on Automated Vocal Fold Segmentation in Ex Vivo High-Speed Phonation Images
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.37860
   Two-dimensional (2D) intraglottal particle image velocimetry (PIV) of ex vivo phonation experiments quantifies vocal fold medial surface dynamics and the intraglottal flow field.
23. SURE-Tagger: Structured Speech Representations for Compositional Evaluation and Targeted Data Selection
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2610.32198
   Speech conveys linguistic content, speaker characteristics, temporal interactions, paralinguistic cues, and acoustic environments.
24. A Wearable Sensor Array on Conductive Textiles for Body Surface Temperature Mapping
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.37008
   Continuous monitoring of body-surface temperature distributions can provide useful information for wearable health and thermal-condition monitoring.
25. What Does LOL Measure? Hidden Label Shifts in Low-Light Enhancement Benchmarks
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.38121
   LOL-v1 and LOLv2-Real are the standard benchmarks for supervised low-light image enhancement, and several leading methods ship hand-set test-time adjustments with their evaluation scripts.
26. MS-Bench: A Condition-Stratified Multi-Speaker ASR Benchmark
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2610.25824
   Multi-speaker automatic speech recognition (MSASR) has advanced rapidly with end-to-end speech large language models.
27. Placing the Beamspace Window on a Hopping Carrier: Bin Drift, Exact Capture and the Cost of Rounding
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.37067
   A frequency-agile radar with a large digital array moves the spatial frequency of its look direction across the beams of a spatial FFT every time it changes carrier, so a beamspace adaptive processor must decide where to put its window at each hop.
28. The Power Watershed as a Differentiable Segmentation Layer
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.38516
   Can a Watershed transform, the classical tool of morphological segmentation, be placed at the end of a deep network and trained through?
29. SLT 2026 REAL-TSE Challenge: Real-world Target Speaker Extraction from Conversational Recordings
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2607.15198
   We introduce the REAL-TSE Challenge, an IEEE SLT 2026 satellite challenge on target speaker extraction~(TSE) from real conversational recordings.
30. Orthogonal Zadoff-Chu Based Pilot Design For Fractional Channel Estimation in MIMO-OTFS
   Source: eess.SP (Signal Processing)
   Link: https://arxiv.org/abs/2610.37079
   In this paper, we propose a Zadoff-Chu (ZC) sequence-based pilot design for fractional channel estimation in orthogonal time frequency space (OTFS) modulation for point-to-point multi-input multi-output (MIMO) scenarios.
31. Enhanced image forensics through partial C2PA-based provenance information
   Source: eess.IV (Image and Video Processing)
   Link: https://arxiv.org/abs/2610.38639
   Cryptographically verifiable C2PA-based Content Credentials (CCs) can be attached to freshly acquired or generated digital images but also to already existing ones with no previous provenance information, for instance after processing through trusted C2PA-enabled software tools.
32. Correspondence Is Not Transfer: A Pairing-Controlled Audit of Clinical Audio-Metadata Alignment
   Source: eess.AS (Audio and Speech Processing)
   Link: https://arxiv.org/abs/2610.35747
   Clinical audio models increasingly align recordings with participant metadata or metadata-derived text.

Sources in this brief: eess.AS (Audio and Speech Processing); eess.IV (Image and Video Processing); eess.SP (Signal Processing).
Selected 32 of 438 available items for this weekly brief.