A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Improved text summarization using belief propagation on weighted bipartite graphs.
ZeroSCROLLS benchmarks zero-shot natural language understanding over long texts.
Adaptive contrastive search improves text generation quality and diversity.
This work uses image generation models to find vision model bugs.
Paper introduces a specialized text classification system for French Open Banking transactions.
MoleculeSTM learns from molecule structures and texts for better drug design.
FinGPT democratizes financial data for LLMs, enabling innovation.
In the past few years, neural abstractive text summarization with sequence-to-sequence (seq2seq) models have gained a lot of popularity. Many interesting techniques have been proposed to improve seq2seq models, making them capable of handling different challenges, such as saliency, fluency and human readability, and ge…
Top-H decoding improves text generation by balancing creativity and coherence.
Inductive transfer learning has greatly impacted computer vision, but existing approaches in NLP still require task-specific modifications and training from scratch. We propose Universal Language Model Fine-tuning (ULMFiT), an effective transfer learning method that can be applied to any task in NLP, and introduce tech…
Open-FinLLMs tackle financial tasks with multimodal capabilities.
Recent work has shown how to train Convolutional Neural Networks (CNNs) rapidly on large image datasets, then transfer the knowledge gained from these models to a variety of tasks. Following [Radford 2017], in this work, we demonstrate similar scalability and transfer for Recurrent Neural Networks (RNNs) for Natural La…
Study examines how decoding algorithms affect fairness in language generation models.
Unified pipeline predicts equipment anomalies with high precision and reduced false positives.
Paper semantifies bioassay text using neural networks.
Text embedding representing natural language documents in a semantic vector space can be used for document retrieval using nearest neighbor lookup. In order to study the feasibility of neural models specialized for retrieval in a semantically meaningful way, we suggest the use of the Stanford Question Answering Dataset…
Humor is a unique and creative communicative behavior displayed during social interactions. It is produced in a multimodal manner, through the usage of words (text), gestures (vision) and prosodic cues (acoustic). Understanding humor from these three modalities falls within boundaries of multimodal language; a recent r…
Hierarchical NMF organizes COVID-19 literature into a searchable tree.
New text-to-image diffusion models improve scene understanding for AI agents.
This research discovers model architecture and training dataset characteristics through strategic input probing.
Generative model evaluates text emotion intensity, outperforming classification.
Mitigates bias in text classification by weighting instances.
SCROLLS benchmarks long text NLP tasks, improving existing models.
New method predicts nonfactuality in LLM responses using semantic isotropy.
MaskGAN fills in text blanks more flexibly and effectively.
Study shows density of mapping classes on infinite-type surfaces using quasi-conformal maps.
Self-supervised learning helps train deep features without needing lots of labeled data.
Deep neural networks are data hungry models and thus face difficulties when attempting to train on small text datasets. Transfer learning is a potential solution but their effectiveness in the text domain is not as explored as in areas such as image analysis. In this paper, we study the problem of transfer learning for…
This paper has two purposes. First it partially extends the result in the author's previous work concerning the asymptotic expansion of the Tian-Yau metrics, by considering a slightly larger class of quasi-projective manifolds. This text is also intended to provide a quick introductory reference to the study of Ricci-f…
A recently introduced text classifier, called SS3, has obtained state-of-the-art performance on the CLEF's eRisk tasks. SS3 was created to deal with risk detection over text streams and, therefore, not only supports incremental training and classification but also can visually explain its rationale. However, little att…
Combines BERT and graph CNN for improved text classification.
EmTract extracts emotions from financial social media text.
RNNs classify text by accumulating evidence on a low-dimensional manifold.
We present RUSLAN -- a new open Russian spoken language corpus for the text-to-speech task. RUSLAN contains 22200 audio samples with text annotations -- more than 31 hours of high-quality speech of one person -- being the largest annotated Russian corpus in terms of speech duration for a single speaker. We trained an e…
Benchmarking AutoML for tables with text fields, achieving top performance.
Open-domain dialog generation is a challenging problem; maximum likelihood training can lead to repetitive outputs, models have difficulty tracking long-term conversational goals, and training on standard movie or online datasets may lead to the generation of inappropriate, biased, or offensive text. Reinforcement Lear…
Estimates watermarked content proportions in mixed-source texts.
A great variety of text tasks such as topic or spam identification, user profiling, and sentiment analysis can be posed as a supervised learning problem and tackle using a text classifier. A text classifier consists of several subprocesses, some of them are general enough to be applied to any supervised learning proble…
Dataset distillation is a method for reducing dataset sizes by learning a small number of synthetic samples containing all the information of a large dataset. This has several benefits like speeding up model training, reducing energy consumption, and reducing required storage space. Currently, each synthetic sample is …
Representing entities and relations in an embedding space is a well-studied approach for machine learning on relational data. Existing approaches, however, primarily focus on simple link structure between a finite set of entities, ignoring the variety of data types that are often used in knowledge bases, such as text, …
Enhances GNNs with text features for better fake news detection.
Today, the dominant paradigm for training neural networks involves minimizing task loss on a large dataset. Using world knowledge to inform a model, and yet retain the ability to perform end-to-end training remains an open question. In this paper, we present a novel framework for introducing declarative knowledge to ne…
Text to Image Synthesis refers to the process of automatic generation of a photo-realistic image starting from a given text and is revolutionizing many real-world applications. In order to perform such process it is necessary to exploit datasets containing captioned images, meaning that each image is associated with on…
Concept Relation Discovery and Innovation Enabling Technology (CORDIET), is a toolbox for gaining new knowledge from unstructured text data. At the core of CORDIET is the C-K theory which captures the essential elements of innovation. The tool uses Formal Concept Analysis (FCA), Emergent Self Organizing Maps (ESOM) and…
Trajectory data suffices for efficient RL in linear MDPs.
Text clustering method replaces centroids with summaries for interpretability and scalability.
LexNLP is an open source Python package focused on natural language processing and machine learning for legal and regulatory text. The package includes functionality to (i) segment documents, (ii) identify key text such as titles and section headings, (iii) extract over eighteen types of structured information like dis…
Paper defends LSTM-based text classification models from backdoor attacks.