Da Tang

Zheng Zhang, Chen Zheng, Da Tang et al.

This paper introduces a multifaceted methodology for fine-tuning and evaluating large language models (LLMs) for specialized monetization tasks. The goal is to balance general language proficiency with domain-specific skills. The methodology has three main components: 1) Carefully blending in-domain and general-purpose data during fine-tuning to achieve an optimal balance between general and specialized capabilities; 2) Designing a comprehensive evaluation framework with 45 questions tailored to assess performance on functionally relevant dimensions like reliability, consistency, and business impact; 3) Analyzing how model size and continual training influence metrics to guide efficient resource allocation during fine-tuning. The paper details the design, data collection, analytical techniques, and results validating the proposed frameworks. It aims to provide businesses and researchers with actionable insights on effectively adapting LLMs for specialized contexts. We also intend to make public the comprehensive evaluation framework, which includes the 45 tailored questions and their respective scoring guidelines, to foster transparency and collaboration in adapting LLMs for specialized tasks.

Jiyao Wang, Qingyong Hu, Duoxun Tang et al.

Video-based remote physiological measurement (RPM) is highly accessible but remains fragile under varying illumination, skin tones, and motion. Radio frequency (RF) radar is largely invariant to illumination and appearance, providing complementary cardio-respiratory micro-motion cues; however, requiring radar at inference is often impractical due to its limited ubiquity and deployment overhead. We propose RPM-Distill, a physiology-guided cross-modal distillation framework that leverages synchronized radar only during training while retaining video-only inference. Our key observation is that although RGB and RF waveforms differ in sensing physics and time-domain morphology, they share similar latent periodic rhythm in the frequency domain. We thus distill physiology-structured spectral evidence to improve robustness, via losses that (i) anchor the fundamental peak, (ii) match the off-peak background distribution, and (iii) preserve spectral morphology and sharpness. To avoid negative transfer under sample-level teacher quality and alignment uncertainty, a spectral policy network predicts sample-level distillation gates and component weights from the student--teacher spectral relation map, learned with a meta bilevel objective on a small labeled validation split. Through extensive experiments in challenging conditions and cross-dataset settings, RPM-Distill brings 81\% MAE and 21\% correlation improvement over unimodal baselines. Code is at https://github.com/WJULYW/RPM-Distill.

Cesar Diaz-Londono, Liu Zhang, Jorge De La Cruz et al.

The rapid growth of electric vehicles (EVs) is increasing the need to accurately quantify their flexibility as a resource for power system operation. However, most existing approaches rely on simplified or power-controllable models that overlook the intrinsic constraints of fast and ultra-fast DC charging. In practice, flexibility is fundamentally shaped by battery management system (BMS) behavior, connection time availability, and battery-protection limits. This paper introduces a trajectory-aware data-driven framework to quantify EV charging flexibility as an energy-bounded and time-constrained process. Based on 252 real charging sessions, 141 representative Power-SoC profiles are reconstructed to capture real-world charging dynamics. Unidirectional flexibility is defined through bounds on the maximum shiftable charging energy, while bidirectional flexibility is quantified as the bounds of the maximum extractable discharge energy under feasibility constraints. Results show that flexibility depends on charging state and connection time. Charging beyond 80% SoC increases duration with limited gains, while higher charger power saturates due to BMS limits. Charging time in the 20%-80% range drops by over 60%, and mean power increases by up to 40%. The maximum extractable bidirectional energy can exceed twice its value depending on the point at which flexibility is activated. These results highlight that EV flexibility is not a controllable resource, but a bounded and time-dependent capability. As such, the proposed framework provides actionable limits that can be directly used by system operators and aggregators for scheduling, peak shaving, and short-duration flexibility services.