Tag
HuggingModels 分享了一个基于 RoBERTa 的模型，可自动对漏洞报告的严重程度进行分类，训练数据来自真实的漏洞评分，旨在帮助安全团队更高效地处理海量 CVE。
该研究提出一种分解式关系抽取框架，在NTSB航空事故叙述基准上评估TF-IDF SVM与RoBERTa在关系稀疏条件下的表现，发现穷举评估下性能大幅下降且假阳性泛滥，凸显候选空间管理与假阳性控制的挑战。
该研究利用RoBERTa掩码语言建模，在科幻与一般小说语料库中对man/woman/person进行掩码替换分析，发现两个语料库都存在跨性别二元的替换，但科幻展现出更强的性别可渗透性与去性别化倾向，为计算文学研究提供了可迁移的分析管线。
作者将两个开源 AI 内容检测模型（RoBERTa 文本检测器约 500MB 压缩至 97MB，以及 26MB 的 ViT 图像检测器）通过 ONNX Runtime Web 和混合精度量化等技术实现在浏览器端完全本地运行，无需上传数据，文本检测约 60ms、图像约 200ms。
This arXiv paper shows that RoBERTa-based machine-text detectors track statistical complexity rather than machine authorship: "humanizing" LLM outputs via Mistral-7B actually raises Verb Diversity and detection scores, while the detector exhibits a 76.3% false-positive rate on formal human writing and is evaded by 41% of homoglyph attacks. Event-based latent-space detection proved fragile to paraphrasing and character-level perturbations.
This paper presents the YNU-HPCC team's approach for SemEval-2025 Task 11 on text-based emotion recognition, using a RoBERTa model with enhanced output headers and achieving a ranking score of 0.44 through English-translated datasets.
A comparative study of BART, BERT, and RoBERTa for text summarization, examining their architectures and suitability for extractive and abstractive summarization tasks.
This paper compares RoBERTa-based sentiment analysis with an LLM-based multi-dimensional framing analysis on political news articles, finding that traditional SA suffers from 'neutral collapse' and that LLM-based approaches better capture bias, sensationalism, and framing for social science research.
Proposes a Hierarchy-Aware RoBERTa framework for classifying cybersecurity vulnerabilities in the CWE taxonomy, demonstrating that hierarchy-aware representation learning is more effective than oversampling techniques for handling class imbalance.
This paper presents UCSC NLP's systems for SemEval-2026 Task 10 (PsyCoMark), addressing conspiracy marker extraction using boundary-aware span extraction with RoBERTa, and document-level conspiracy classification with label smoothing. The systems ranked 7th in subtask 1 and 12th in subtask 2.
This paper proposes a dual-threshold hard example mining strategy for cross-platform Chinese offensive comment detection, addressing performance degradation due to domain shift. The method fine-tunes a RoBERTa model on the COLD dataset and adapts it to four Chinese social media platforms with minimal labeled data.
HierBias introduces a hierarchical context-conditioned model for media bias detection that leverages document context to improve sentence-level classification, achieving state-of-the-art F1 and MCC on the BABE and BASIL datasets.
A fine-grained study of narrative features in web-scale LLM pretraining data, introducing NarraBERT and NarraDolma to measure narrative patterns and their distribution across sources.
This paper introduces a controlled content overlap setup using parallel Bible translations to evaluate how much style classifiers rely on content cues rather than actual style features. Results show that low-overlap models degrade when content cues are removed, while high-overlap models transfer more robustly.
This paper demonstrates that fine-tuned AI text detectors amplify a pretrained typicality axis rather than learning an AI-vs-human boundary, with raw encoder projections often matching or exceeding fine-tuned performance.
ETH Zurich researchers show that fine-tuned RoBERTa models can infer users’ Big-Five personality traits from ChatGPT chat logs with up to 44 % above-random accuracy, highlighting privacy risks of conversational AI.