On-device research index

arXiv research

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.

169,051 papers · 148 categories

Trend · papers per month

count % of corpus

6 results for “WFSTs”

Differentiable Weighted Finite-State Transducerscs.LG

Framework for differentiating WFSTs for structured loss functions.

problem Training dynamic structured loss functions in neural networks.

method Automatic differentiation with WFSTs, combining pruning and back-off.

result Demonstrated learning over WFST latent phrase decomposition.

differentiation finite-state transducers structured loss neural networks pruning

2020-10-02abs ↗pdf ↗

Star Temporal Classification: Sequence Classification with Partially Labeled Datacs.LG

Develops STC for sequential data with missing labels.

problem Learning from partially labeled and unsegmented sequential data.

method Introduces Star Temporal Classification (STC) using a star token and GTN framework.

result Recover most of supervised baseline performance with up to 70% missing labels.

Sequential Data Partial Labels Star Token Finite-State Transducers Automatic Differentiation

2022-01-28abs ↗pdf ↗

ELICA: An Automated Tool for Dynamic Extraction of Requirements Relevant Informationcs.SE

ELICA helps analysts understand unfamiliar domains by extracting relevant terms.

problem Communication barriers between analysts and stakeholders in unfamiliar domains.

method ELICA uses WFSTs to dynamically extract and label requirements-relevant knowledge from text and non-linguistic cues.

result ELICA supports analysts in understanding and eliciting requirements from unfamiliar domains.

requirements elicitation dynamic extraction natural language processing WFSTs domain understanding

2018-07-21abs ↗pdf ↗

Dynamic Visual Analytics for Elicitation Meetings with ELICAcs.CY

ELICA helps analysts extract relevant information during elicitation meetings.

problem Challenges in eliciting requirements due to analyst's lack of domain knowledge.

method ELICA uses a novel information extraction algorithm combining WFSTs and SVMs, presented in an interactive GUI.

result ELICA effectively extracts relevant information in real-time and facilitates interactive and dynamic process.

requirements elicitation interactive visual analytics information extraction machine learning domain knowledge

2018-07-10abs ↗pdf ↗

A Comparison of Hybrid and End-to-End Models for Syllable Recognitioncs.CL

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.

syllable recognition end-to-end models hybrid models German Verbmobil corpus

2019-09-19abs ↗pdf ↗

Learning Sharing Behaviors with Arbitrary Numbers of Agentscs.LG

Modeling and learning turn-taking behaviors in multi-agent systems.

problem Modeling and predicting turn-taking behaviors in dynamic multi-agent systems.

method Individual behavior models (WFSTs) and multi-agent fusion model (logistic regression classifier).

result Accurately models and predicts turn-taking behaviors with high precision.

multi-agent systems turn-taking behaviors Q-learning logistic regression finite state transducers

2018-12-10abs ↗pdf ↗