German FinBERT improves financial text analysis performance.
problem Capturing domain-specific nuances in financial text.
method German FinBERT is a pre-trained German language model trained on a large corpus of financial data.
result German FinBERT outperforms standard models on finance-specific tasks.
Improved neural NER by optimizing large corpora for German.
problem Low-resource language named entity recognition.
method Optimized large corpora, lemmatization, part-of-speech tagging, and detailed optimization.
result Up to 11% improvement in F-score on German NER tasks.
Improved neural language models using adversarial training.
problem Overfitting in large-scale neural language models.
method Adversarial training mechanism to introduce adversarial noise to output embedding layers.
result Improved perplexity scores and BLEU scores on various language tasks.
Evolved Transformer improves on Transformer architecture for language tasks.
problem Improving Transformer architecture for sequence tasks.
method Evolutionary architecture search with warm starting and dynamic resource allocation.
result Evolved Transformer achieves state-of-the-art BLEU scores and reduces parameter count.
Study finds common poetic themes across languages over time.
problem Understanding thematic evolution in different poetic traditions.
method Applied Latent Dirichlet Allocation (LDA) to poetry corpora of four languages.
result Identified common themes and their temporal trends across poetic traditions.
We describe a simple neural language model that relies only on character-level inputs. Predictions are still made at the word-level. Our model employs a convolutional neural network (CNN) and a highway network over characters, whose output is given to a long short-term memory (LSTM) recurrent neural network language mo…
In this paper, we present Neural Phrase-based Machine Translation (NPMT). Our method explicitly models the phrase structures in output sequences using Sleep-WAke Networks (SWAN), a recently proposed segmentation-based sequence modeling method. To mitigate the monotonic alignment requirement of SWAN, we introduce a new …
Hybrid and end-to-end models compare in syllable recognition.
problem Comparing hybrid and end-to-end models for syllable recognition.
method Traditional hybrid system (kaldi) vs. end-to-end (TensorFlow) models.
result Hybrid models with explicit syllable knowledge outperform end-to-end models.
New models improve morpheme segmentation in low-resource languages.
problem Improving morpheme segmentation in low-resource languages.
method Two new models: LSTM pointer-generator and sequence-to-sequence with hard monotonic attention.
result Novel models outperform existing ones by up to 11.4% accuracy in low-resource settings.
End-to-end training of automated speech recognition (ASR) systems requires massive data and compute resources. We explore transfer learning based on model adaptation as an approach for training ASR models under constrained GPU memory, throughput and training data. We conduct several systematic experiments adapting a Wa…
Cross-language learning allows us to use training data from one language to build models for a different language. Many approaches to bilingual learning require that we have word-level alignment of sentences from parallel corpora. In this work we explore the use of autoencoder-based methods for cross-language learning …
Study examines bias in language models across multiple languages.
problem Assessing bias in language models across different languages.
method Semi-automatically translated data sets into multiple languages, analyzed mono- and multilingual models.
result Notable differences in bias across languages, with Turkish models showing least stereotypes.
Neural Machine Translation (MT) has reached state-of-the-art results. However, one of the main challenges that neural MT still faces is dealing with very large vocabularies and morphologically rich languages. In this paper, we propose a neural MT system using character-based embeddings in combination with convolutional…
MGLM models all possible language channel factorizations for improved multilingual generation.
problem Generating multilingual text with flexibility and quality.
method Generative joint distribution model over language channels, marginalizing all possible factorizations.
result MGLM outperforms traditional models in multilingual generation tasks.
Bayesian SHMM discovers acoustic units from unlabeled speech.
problem Discovering language-specific acoustic units from unlabeled speech.
method Bayesian Subspace Hidden Markov Model (SHMM) trained on labeled data to find new acoustic units on target language.
result Significantly outperforms previous HMM-based systems and compares favorably with Variational Auto Encoder-HMM.
SSMBA generates synthetic data to improve robustness in natural language tasks.
problem Improving out-of-domain generalization of models trained on natural language data.
method SSMBA uses corruption and reconstruction functions to generate synthetic data points on the manifold assumption.
result SSMBA consistently outperforms existing methods on robustness benchmarks across multiple tasks and datasets.
Proposes a regularization approach to model German power derivative market, identifying significant risk spillovers.
problem Large portfolio of German power derivative contracts, identifying significant risk spillovers.
method Combines high-dimensional variable selection with dynamic network analysis.
result Identifies significant risk contributors and interdependencies between contracts, especially spot contracts.
In this paper, we attempt to improve Statistical Machine Translation (SMT) systems on a very diverse set of language pairs (in both directions): Czech - English, Vietnamese - English, French - English and German - English. To accomplish this, we performed translation model training, created adaptations of training sett…
Unified BERT model improves NER across multiple languages.
problem Language-specific NER models limit data extraction.
method Jointly trained multilingual BERT with regularization.
result Unified model outperforms monolingual models on various datasets.
Paper classifies Parkinson's disease from speech in three languages using CNNs and transfer learning.
problem Classifying Parkinson's disease from speech in multiple languages.
method Convolutional Neural Networks (CNNs) with transfer learning among Spanish, German, and Czech.
result Transfer learning improves model accuracy by up to 8% and balances specificity-sensitivity.
This research predicts stock market movements using Vision-Language models.
problem Predicting future stock market direction using historical data.
method Utilizing image and byte-based representations of stock data processed with Vision-Language models.
result The proposed approach significantly outperforms deep learning baselines.
Study uses few-shot learning to analyze claims and arguments in German debate on arms deliveries.
problem Limited data and computational resources for automated content analysis.
method Multilingual transformer model with adapter extension and few-shot learning.
result Parameter-efficient approach performs well on varying training set sizes.
Study reveals stylized facts in German bond futures markets.
problem Understanding market dynamics in German bond futures.
method Analyzed tick-by-tick data of four German bond futures contracts.
result Uncovered commonalities and unique characteristics across different futures.
LLM Pro Finance Suite enhances financial NLP with instruction-tuned models.
problem Limited NLP capabilities for financial tasks in generalist models.
method Instruction-tuned large language models fine-tuned on financial data.
result Consistent improvement over state-of-the-art baselines in finance tasks.
We introduce a new approach to unsupervised estimation of feature-rich semantic role labeling models. Our model consists of two components: (1) an encoding component: a semantic role labeling model which predicts roles given a rich set of syntactic and lexical features; (2) a reconstruction component: a tensor factoriz…
Soft diamond regularizers improve deep learning performance and sparsity.
problem Improving deep learning performance and sparsity of trained weights.
method New soft diamond synaptic weight priors based on thick-tailed symmetric alpha stable probability curves.
result Soft diamond regularizers outperform state-of-the-art methods in deep learning tasks.
The paper analyzes and forecasts intraday electricity prices using econometric models.
problem Analyzing and forecasting the efficiency of the German Intraday Continuous electricity market.
method Multivariate econometric time series model with lasso and elastic net techniques.
result The model provides new insights into the ID3-Price behavior and market efficiency. How does dynamic price information flow among Northern European electricity spot prices and prices of major electricity generation fuel sources? We use time series models combined with new advances in causal inference to answer these questions. Applying our methods to weekly Nordic and German electricity prices, and oi…
Syntax-enhanced models boost machine translation and NLP performance.
problem Limited training data and complex models struggle in NLP tasks.
method Syntax information was explicitly fed into Transformer and BERT models.
result Syntax-infused models achieved significant BLEU improvements.
We present a relatively detailed analysis of the persistence probability distributions in financial dynamics. Compared with the auto-correlation function, the persistence probability distributions describe dynamic correlations non-local in time. Universal and non-universal behaviors of the German DAX and Shanghai Index…
Short-term probabilistic forecasting of German electricity imbalance prices.
problem Uncertainty in renewable energy capacity and electricity prices.
method Combining lasso with bootstrap, gamlss, and probabilistic neural networks for forecasting imbalance prices.
result Sophisticated methods improve empirical coverage of imbalance prices but do not substantially outperform the intraday continuous price index.
Detects out-of-distribution sentences in Neural Machine Translation.
problem Identifying sentences from a different language than the training data.
method Developed a new uncertainty measure for long sequences of words in Transformers.
result Shows ability to identify Dutch sentences as German input.
Research optimizes a small RES utility's portfolio by dynamically trading in German electricity markets.
problem Managing risks in RES producers and electricity traders in changing electricity markets.
method Uses SVAR model to estimate market relationships and data-driven trading strategies to optimize revenue and reduce risk.
result Data-driven trading strategies increase utility revenue and reduce trading risk.
Investigates if adding cryptocurrencies to German portfolios diversifies better, finding mixed results.
problem Improving diversification in German investor portfolios using cryptocurrencies.
method Portfolio analysis with descriptive statistics, graphical methods, and econometric spanning tests, using a customized EWCI.
result Cryptocurrencies can improve diversification in some windows but not as a normal case.
The grid integration of intermittent Renewable Energy Sources (RES) causes costs for grid operators due to forecast uncertainty and the resulting production schedule mismatches. These so-called profile service costs are marginal cost components and can be understood as an insurance fee against RES production schedule u…
Paper examines NMT robustness to nonsensical inputs.
problem NMT systems fail when source sentences are altered.
method Soft-attention technique to replace words in source sentences.
result Proposed technique achieves high success rate and outperforms existing methods.
Low redispatch prices boost green hydrogen production cost, encouraging electrolyzer siting.
problem Uncertainty in redispatch power availability and its impact on green hydrogen production cost.
method Historic redispatch time series analysis and power purchase scenarios evaluation.
result Low price levels can lead to notable production cost reductions, incentivizing electrolyzer siting.
Unified approach for generating sequences from undirected models.
problem Generating sequences directly from undirected models like BERT.
method Generalized model of sequence generation unifying decoding in directed and undirected models.
result Adapted decoding algorithms for undirected models achieve competitive results.
Based on the tick-by-tick stock prices from the German and American stock markets, we study the statistical properties of the distribution of the individual stocks and the index returns in highly collective and noisy intervals of trading, separately. We show that periods characterized by the strong inter-stock coupling…
Study on excess mortality in Germany during 2020-21.
problem Analyzing excess mortality during the pandemic in Germany.
method Empirical study using official death counts.
result Provided conclusions for insurance businesses.
For the first time, we apply the wavelet coherence methodology on biofuels (ethanol and biodiesel) and a wide range of related commodities (gasoline, diesel, crude oil, corn, wheat, soybeans, sugarcane and rapeseed oil). This way, we are able to investigate dynamics of correlations in time and across scales (frequencie…
Lognormal distribution used for predicting team rankings in an orienteering relay race.
problem Predicting final team rankings in an orienteering relay race.
method Used lognormal distribution and Fenton-Wilkinson approximations for order statistics.
result Accurate predictions of team rankings using order statistics.
Paper speeds up neural language model inference by 20x for top-k word prediction.
problem Slow inference speed of neural language models on mobile devices.
method Introduced a screening model using Gumbel softmax to approximate softmax layer.
result Achieved 20.4x speedup with 98.9% precision@1 and 99.3% precision@5 for German to English translation.
We investigate the large-volatility dynamics in financial markets, based on the minute-to-minute and daily data of the Chinese Indices and German DAX. The dynamic relaxation both before and after large volatilities is characterized by a power law, and the exponents p± usually vary with the strength of the large vo…
We investigate the large-fluctuation dynamics in financial markets, based on the minute-to-minute and daily data of the Chinese Indices and German DAX. The dynamic relaxation both before and after the large fluctuations is characterized by a power law, and the exponents p± usually vary with the strength of the lar…
HighD dataset captures naturalistic vehicle behavior on German highways for automated driving validation.
problem Current measurement methods fail to meet requirements for scenario-based validation of highly automated vehicles.
method Aerial perspective data collection fulfilling naturalistic behavior, quantity, and variety requirements.
result 16.5 hours of measurements from six locations, 110,000 vehicles, 45,000 km driven, 5600 lane changes.
Company2Vec creates embeddings from company websites for fine-grained business analytics.
problem Lack of fine-grained company labels for analytics.
method Word2Vec and dimensionality reduction on company website data.
result Semantic company embeddings for various applications.
Study shows German day-ahead spot market becoming less volatile.
problem Intermittent renewable energy sources increase price volatility.
method Applied singular value decomposition (SVD) for volatility measurement.
result Day-ahead market is becoming less volatile over time.