Tag
Yale researchers introduce the Parallel-Then-Integrated (PTI) procedure, a four-step method for integrating two computational representations of qualitative text—ontology-guided PVminer coding and NMF-derived themes—demonstrated on interview data from Nepalese mothers.
This paper replicates a distributional-semantics extractive summarization method for Hindi and evaluates it on standard corpora, finding that sentence position is the only contributing feature and current Hindi benchmarks fail to incentivize advanced content selection.
This paper demonstrates that large language models (LLMs) achieve annotation quality equivalent to human experts, suggesting the main challenge is minimizing ambiguity in coding rules rather than choosing between human and machine coders.
This paper proposes a neural network model for fast and accurate identification of text content file types, outperforming existing tools like Magika in accuracy and speed while being smaller in size.
This paper introduces hierarchical rotary positional encoding (hRoPE) to represent text hierarchy and uses compression depth as a signature of paragraph structure, showing it correlates with genuine hierarchical organization better than other metrics.
This paper demonstrates that removing stopwords in legal text analysis distorts doctrinal and ideological signals, recommending that stopwords be retained for measurement validity.
Pangram is an AI detector tool that accurately identifies AI-generated text and images, validated by third-party research and used by institutions worldwide.
Introduces DP-SPIN, a differentially private framework for aggregate measurement and summarization over semantic concepts in text data, with evaluations on CFPB, Amazon, and Yelp datasets.
The paper proposes a training-free and alignment-free approach to corporate intelligence using deterministic sparse seed vectors for analyzing SEC filings, enabling efficient document comparison and semantic event detection without large language models.
MonoTM is an interpretable topic modeling framework that decouples mixture estimation from feature interpretation using sparse autoencoders, providing semantically meaningful topics beyond traditional word-based representations.
The article explores Pangram, an AI startup that detects AI-generated text, its role in publishing scandals, and questions about the trustworthiness of its detection model.
This paper presents a benchmark for classifying research papers into good (highly cited) and non-good (retracted) using textual features from titles and abstracts, with models like SVM and neural networks achieving up to 91.12% accuracy.
BERTilda is an explainable framework that tracks topic lifecycles in longitudinal text streams by constructing temporal graphs with similarity and flow signals to detect splits, merges, and other transitions, achieving high agreement rates on annotated datasets.
Claude Watermark is a tool designed to detect and remove AI-generated watermarks from text, helping users manage AI traces for privacy or authenticity.
Ancient Library is an online reader offering 1,060 Greek and Latin classics where clicking any word displays its lemma, morphology, and full dictionary entries.
This study applies computational linguistics and supervised machine learning to predict early-stage startup exits from textual descriptors alone, finding that founder narratives carry predictive signal and introducing a quantifiable Hyping Score.
A web-based tool that detects and highlights common clichéd phrases in LLM-generated text, such as 'no X, no Y' chains and 'sit with that', to help identify AI writing patterns.
This paper proposes two novel normalisation techniques (Square Root Correction and Hapax Correction) for deriving likelihood ratios from the LambdaG authorship verification method without needing a separate calibration model, reducing data requirements and complexity while achieving comparable or superior performance in forensic text comparison.
This paper proposes a multidimensional text analysis approach combining Japanese NLP metrics and statistical methods to evaluate changes in risk disclosure quality, applied to Japan's 2019 corporate disclosure reforms. The analysis of 19,770 firm-year observations reveals complex shifts such as increased volume accompanied by decreased readability.
This paper applies computational stylometry to English translations of the Pali Canon, examining vocabulary differences across the Sutta, Vinaya, and Abhidhamma divisions.