Arrow Research search

Author name cluster

Felipe Soares

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers
1 author row

Possible papers

2

NeurIPS Conference 2025 Conference Paper

HelpSteer3-Preference: Open Human-Annotated Preference Data across Diverse Tasks and Languages

  • Zhilin Wang
  • Jiaqi Zeng
  • Olivier Delalleau
  • Hoo-Chang Shin
  • Felipe Soares
  • Alexander Bukharin
  • Ellie Evans
  • Yi Dong

Preference datasets are essential for training general-domain, instruction-following language models with Reinforcement Learning from Human Feedback (RLHF). Each subsequent data release raises expectations for future data collection, meaning there is a constant need to advance the quality and diversity of openly available preference data. To address this need, we introduce HelpSteer3-Preference, a permissively licensed (CC-BY-4. 0), high-quality, human-annotated preference dataset comprising of over 40, 000 samples. These samples span diverse real-world applications of large language models (LLMs), including tasks relating to STEM, coding and multilingual scenarios. Using HelpSteer3-Preference, we train Reward Models (RMs) that achieve top performance on RM-Bench (82. 4%) and JudgeBench (73. 7%). This represents a substantial improvement (~10% absolute) over the previously best-reported results from existing RMs. We demonstrate HelpSteer3-Preference can also be applied to train Generative RMs and how policy models can be aligned with RLHF using our RMs.

PDF Details

AIIM Journal 2017 Journal Article

A hierarchical classifier based on human blood plasma fluorescence for non-invasive colorectal cancer screening

  • Felipe Soares
  • Karin Becker
  • Michel J. Anzanello

Colorectal cancer (CRC) a leading cause of death by cancer, and screening programs for its early identification are at the heart of the increasing survival rates. To motivate population participation, non-invasive, accurate, scalable and cost-effective diagnosis methods are required. Blood fluorescence spectroscopy provides rich information that can be used for cancer identification. The main challenges in analyzing blood fluorescence data for CRC classification are related to its high dimensionality and inherent variability, especially when analyzing a small number of samples. In this paper, we present a hierarchical classification method based on plasma fluorescence to identify not only CRC, but also adenomas and other non-malignant colorectal findings that may require further medical investigation. A feature selection algorithm is proposed to deal with the high dimensionality and select discriminant fluorescence wavelengths. These are used to train a binary support vector machine (SVM) in the first level to identify the CRC samples. The remaining samples are then presented to a one-class SVM trained on healthy subjects to detect deviant samples, and thus non-malignant findings. This hierarchical design, together with the one class-SVM, aims to reduce the effects of small samples and high variability. Using a dataset analyzed in previous studies comprised of 12, 341 wavelengths, we achieved much superior results. Sensitivity and specificity are 0. 87 and 0. 95 for CRC detection, and 0. 60 and 0. 79 for non-malignant findings, respectively. Compared to related work, the proposed method presented a better accuracy, required fewer features, and provides a unified approach that expands CRC detection to non-malignant findings.

Details DOI
v2026.09.13