Tootfinder

Joint ASR and Speaker Role Tagging with Serialized Output Training
Anfeng Xu, Tiantian Feng, Shrikanth Narayanan
https://arxiv.org/abs/2506.10349 https://

Joint ASR and Speaker Role Tagging with Serialized Output Training
Automatic Speech Recognition systems have made significant progress with large-scale pre-trained models. However, most current systems focus solely on transcribing the speech without identifying speaker roles, a function that is critical for conversational AI. In this work, we investigate the use of serialized output training (SOT) for joint ASR and speaker role tagging. By augmenting Whisper with role-specific tokens and fine-tuning it with SOT, we enable the model to generate role-aware trans…

Dynamic Beyond 5G and 6G Connectivity: Leveraging NTN and RIS Synergies for Optimized Coverage and Capacity in High-Density Environments
Valdemar Farr\'e, Juan Estrada, David Vega, Luis F Urquiza-Aguiar, Juan A. V\'asquez Peralvo, Symeon Chatzinotas
https://arxiv.org/abs/2506.10900…

Dynamic Beyond 5G and 6G Connectivity: Leveraging NTN and RIS Synergies for Optimized Coverage and Capacity in High-Density Environments
The increasing demand for reliable, high-capacity communication during large-scale outdoor events poses significant challenges for traditional Terrestrial Networks (TNs), which often struggle to provide consistent coverage in high-density environments. This paper presents a novel 6G radio network planning framework that integrates Non-Terrestrial Networks (NTNs) with Reconfigurable Intelligent Surfaces (RISs) to deliver ubiquitous coverage and enhanced network capacity. Our framework overcomes …

"The world is full of darkness. So much is going wrong. Experts agree that America has succumbed to right-wing authoritarianism; call it fascism or something else, these are extraordinarily difficult times.
This post is a break from all of that. At least kind of.
In this piece, I will try and convince you that Doctor Who is the best TV show ever made, explain to you why it matters, and why it’s particularly important in our current context. In a time when cruelty and fear dominate headlines, it’s worth celebrating a show that insists on the power of kindness, intellect, and hope."
#DoctorWho
https://werd.io/2025/doctor-who-is-the-best-show-ever-made-heres-why

This https://arxiv.org/abs/2409.15104 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csDC_…

PecSched: Preemptive and Efficient Cluster Scheduling for LLM Inference
The scaling of transformer-based Large Language Models (LLMs) has significantly expanded their context lengths, enabling applications where inputs exceed 100K tokens. Our analysis of a recent Azure LLM inference trace reveals a highly skewed long-tail distribution of input lengths, with approximately 80% of inputs shorter than 2K tokens. Long inputs constitute only a small fraction. Existing cluster-level LLM scheduling strategies, including First-In-First-Out (FIFO), reservation-based, and pri…

Late gas released in the young Kuiper belt could have significantly contributed to the carbon enrichment of the atmospheres of Neptune and Uranus
Paul Huet, Quentin Kral, Tristan Guillot
https://arxiv.org/abs/2506.03412

Late gas released in the young Kuiper belt could have significantly contributed to the carbon enrichment of the atmospheres of Neptune and Uranus
Exo-Kuiper belts have been observed for decades, but the recent detection of gas in some of them may change our view of the Solar System's youth. Late gas produced by the sublimation of CO (or CO$_2$) ices after the dissipation of the primordial gas could be the norm in young planetesimal belts. Hence, a gas-rich Kuiper belt could have been present in the Solar System. The high C/H ratios observed on Uranus and Neptune could be a clue to the existence of such late gas that could have been accre…

This https://arxiv.org/abs/2505.19746 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csCV_…

Improving Heart Rejection Detection in XPCI Images Using Synthetic Data Augmentation
Accurate identification of acute cellular rejection (ACR) in endomyocardial biopsies is essential for effective management of heart transplant patients. However, the rarity of high-grade rejection cases (3R) presents a significant challenge for training robust deep learning models. This work addresses the class imbalance problem by leveraging synthetic data generation using StyleGAN to augment the limited number of real 3R images. Prior to GAN training, histogram equalization was applied to sta…

A Reduced Cost Two-component Relativistic Equation-of-Motion Coupled Cluster Method for Ionization Potential
Somesh Chamoli, Malaya K. Nayak, Achintya Kumar Dutta
https://arxiv.org/abs/2506.06805

A Reduced Cost Two-component Relativistic Equation-of-Motion Coupled Cluster Method for Ionization Potential
We report an efficient implementation of the ionization potential (IP) variant of the equation-of-motion coupled cluster (IP-EOM-CC) method based on the exact two-component atomic mean field (X2CAMF) framework, utilizing Cholesky decomposition (CD) and frozen natural spinors (FNS). The CD approximation significantly reduces memory demands, whereas the FNS approximation lowers the number of floating-point operations. Together, these techniques make the method computationally efficient for accura…

Lightweight Prompt Biasing for Contextualized End-to-End ASR Systems
Bo Ren, Yu Shi, Jinyu Li
https://arxiv.org/abs/2506.06252 https://

Lightweight Prompt Biasing for Contextualized End-to-End ASR Systems
End-to-End Automatic Speech Recognition (ASR) has advanced significantly yet still struggles with rare and domain-specific entities. This paper introduces a simple yet efficient prompt-based biasing technique for contextualized ASR, enhancing recognition accuracy by leverage a unified multitask learning framework. The approach comprises two key components: a prompt biasing model which is trained to determine when to focus on entities in prompt, and a entity filtering mechanism which efficiently…

Study of electronic band alignment in SiGeSn/GeSn quantum well via internal photoemission effect
Justin Rudie, Huong Tran, Yang Zhang, Sylvester Amoah, Sudip Acharya, Hryhorii Stanchu, Mansour Mortazavi, Timothy A. Morgan, Gregory T. Forcherio, Greg Sun, Gregory Salamo, Wei Du, Shui-Qing Yu
https://arxiv.org/abs/2506.06976

Study of electronic band alignment in SiGeSn/GeSn quantum well via internal photoemission effect
SiGeSn-based optoelectronic devices, which operate across a broad infrared wavelength range, have attracted significant attention, particularly heterostructures utilizing quantum wells are widely utilized. In these structures, band alignment type and barrier height are crucial for carrier confinement, making them highly desirable information to obtain. This work leverages the internal photoemission effect to extract effective barrier heights from a Si0.024Ge0.892Sn0.084 / Ge0.882Sn0.118 single …