Search for: All records

Finetuned large language models (LLMs) have shown remarkable performance in financial tasks, such as sentiment analysis and information retrieval. Due to privacy concerns, finetuning and deploying financial LLMs (FinLLMs) locally are crucial for institutions and individuals. In this paper, we employ quantized low-rank adaptation (QLoRA) to finetune FinLLMs, which leverage low-rank structure and quantization technique to significantly reduce computational requirements while maintaining model performance. We also employ data and pipeline parallelism to enable local finetuning on commodity GPUs. Experiments on financial datasets validate the efficacy of our approach in yielding notable improvements over the base models. 

Fairness in AI and ML systems is increasingly linked to the proper treatment and recognition of data workers involved in training dataset development. Yet, those who collect and annotate the data, and thus have the most intimate knowledge of its development, are often excluded from critical discussions. This exclusion prevents data annotators, who are domain experts, from contributing effectively to dataset contextualization. Our investigation into the hiring and engagement practices of 52 data work requesters on platforms like Amazon Mechanical Turk reveals a gap: requesters frequently hold naive or unchallenged notions of worker identities and capabilities and rely on ad-hoc qualification tasks that fail to respect the workers' expertise. These practices not only undermine the quality of data but also the ethical standards of AI development. To rectify these issues, we advocate for policy changes to enhance how data annotation tasks are designed and managed and to ensure data workers are treated with the respect they deserve. 

Abstract We observed a terrestrial gamma‐ray flash (TGF) shortly after the return stroke in a positive cloud‐to‐ground (+CG) flash during a 2025 winter thunderstorm in the Hokuriku region of Japan. The event was observed with multiple gamma‐ray detectors and radio antenna systems. We identify several distinctions between our event and typical +CG lightning, including: an inverted tripolar storm charge structure, a peak current (190 kA) much higher than reported averages, shorter‐than‐average leader duration between first radio source and return stroke (3.6 vs. 56 ms), and unusual symmetry in the return stroke RF waveform relating to rise and fall time. Many of these differences are consistent with disparities between usual −CG lightning and energetic compact strokes (ECSs), and may be evidence of a positive‐polarity class of ECS events. In addition, we find our TGF observation to be a distinctly short and bright event among others reported in literature. 

We implement and test transfer learning-based gain models across 16 ROADM EDFAs, which achieve less than 0.17/0.30 dB mean absolute error for booster/pre-amplifier gain prediction using only 0.5% of the full target EDFA dataset.