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

5 results for “KL-Vanishing”

Improve Diverse Text Generation by Self Labeling Conditional Variational Auto Encoderstat.ML

Proposes SLCVAE to improve text diversity by self-labeling.

problem KL-Vanishing problem in CVAE for diverse text generation.

method Explicit optimizing objective to guide encoder towards best encoder, using a labeling network.

result Improves text diversity while maintaining comparable accuracy.

text generation diversity conditional variational autoencoder KL-Vanishing self-labeling

2019-03-26abs ↗pdf ↗

Cyclical Annealing Schedule: A Simple Approach to Mitigating KL Vanishingcs.LG

Cyclical Annealing Schedule improves VAE performance in NLP tasks.

problem KL term vanishing in VAEs with auto-regressive decoders.

method Cyclical Annealing Schedule to vary β over multiple cycles.

result Progressive learning of meaningful latent codes improves VAE performance.

VAE KL regularization Auto-regressive decoder Cyclical Annealing NLP tasks

2019-03-25abs ↗pdf ↗

FRAME Revisited: An Interpretation View Based on Particle Evolutioncs.LG

Paper revisits FRAME model, explaining instability and proposing a new metric.

problem Unstable training energy in FRAME model.

method Theoretical analysis using particle physics, proposing a new Wasserstein distance.

result Proposed Wasserstein distance stabilizes energy dissipation and maintains statistical consistency.

FRAME model Maximum Likelihood Estimation Wasserstein distance KL-vanishing issue Particle physics

2018-12-04abs ↗pdf ↗

ControlVAE: Controllable Variational Autoencodercs.LG

ControlVAE improves VAE performance by adding a controller to tune hyperparameters.

problem Existing VAE models struggle with KL vanishing and low reconstruction quality.

method ControlVAE combines a controller inspired by automatic control theory with VAE to improve performance.

result ControlVAE achieves better disentangling and reconstruction quality than existing methods.

VAE disentangling language modeling image generation controller

2020-04-13abs ↗pdf ↗

Stochastic Graph Recurrent Neural Networkstat.ML

SGRNN models evolving graph data for better property prediction.

problem Modeling evolving graph data for property prediction.

method SGRNN uses stochastic latent variables to capture both node attribute and topology evolution, with semi-implicit variational inference and KL-divergence simplification.

result SGRNN improves property prediction on real-world datasets.

graph representation learning stochastic latent variables semi-implicit variational inference evolving graphs property prediction

2020-09-01abs ↗pdf ↗