13 papers
WaveSSM: Multiscale State-Space Models for Non-stationary Signal Attention
Ruben Solozabal, Velibor Bojkovic, Hilal Alquabeh +3
State-space models (SSMs) have emerged as a powerful foundation for long-range sequence modeling, with the HiPPO framework showing that continuous-time projection operators can be…
Uncovering the Spectral Bias in Diagonal State Space Models
Ruben Solozabal, Velibor Bojkovic, Hilal AlQuabeh +2
Current methods for initializing state space models (SSMs) parameters mainly rely on the \textit{HiPPO framework}, which is based on an online approximation of orthogonal polynomia…
SPACT18: Spiking Human Action Recognition Benchmark Dataset with Complementary RGB and Thermal Modalities
Yasser Ashraf, Ahmed Sharshar, Velibor Bojkovic +1
Spike cameras, bio-inspired vision sensors, asynchronously fire spikes by accumulating light intensities at each pixel, offering ultra-high energy efficiency and exceptional tempor…
Mechanistic Insights into Grokking from the Embedding Layer
H. V. AlquBoj, Hilal AlQuabeh, Velibor Bojkovic +2
Grokking, a delayed generalization in neural networks after perfect training performance, has been observed in Transformers and MLPs, but the components driving it remain underexpl…
Number Representations in LLMs: A Computational Parallel to Human Perception
H. V. AlquBoj, Hilal AlQuabeh, Velibor Bojkovic +4
Humans are believed to perceive numbers on a logarithmic mental number line, where smaller values are represented with greater resolution than larger ones. This cognitive bias, sup…
RECALL: Library-Like Behavior In Language Models is Enhanced by Self-Referencing Causal Cycles
Munachiso Nwadike, Zangir Iklassov, Toluwani Aremu +6
We introduce the concept of the self-referencing causal cycle (abbreviated RECALL) - a mechanism that enables large language models (LLMs) to bypass the limitations of unidirection…
A Decade of Deep Learning: A Survey on The Magnificent Seven
Dilshod Azizov, Muhammad Arslan Manzoor, Velibor Bojkovic +9
Deep learning has fundamentally reshaped the landscape of artificial intelligence over the past decade, enabling remarkable achievements across diverse domains. At the heart of the…
A note on rational surgeries on a Hopf link
Velibor Bojković, Jovana Nikolić, Mladen Zekić
It is clear that every rational surgery on a Hopf link in -sphere is a lens space surgery. In this note we give an explicit computation which lens space is a resulting manifold.…
FTBC: Forward Temporal Bias Correction for Optimizing ANN-SNN Conversion
Xiaofeng Wu, Velibor Bojkovic, Bin Gu +2
Spiking Neural Networks (SNNs) offer a promising avenue for energy-efficient computing compared with Artificial Neural Networks (ANNs), closely mirroring biological neural processe…
Energy Efficient Training of SNN using Local Zeroth Order Method
Bhaskar Mukhoty, Velibor Bojkovic, William de Vazelhes +3
Spiking neural networks are becoming increasingly popular for their low energy requirement in real-world tasks with accuracy comparable to the traditional ANNs. SNN training algori…
Metric uniformization of morphisms of Berkovich curves via -adic differential equations
Francesco Baldassarri, Velibor Bojković
We consider a finite étale morphism of quasi-smooth Berkovich curves over a complete nonarchimedean non-trivially valued field , assumed algebraically closed and of…
On the number of connected components of the ramification locus of a morphism of Berkovich curves
Velibor Bojković, Jérôme Poineau
Let be a complete, nontrivially valued non-archimedean field. Given a finite morphism of quasi-smooth -analytic curves that admit finite triangulations, we provide upper bou…
Riemann-Hurwitz formula for finite morphisms of -adic curves
Velibor Bojković
Given a finite morphism of quasi-smooth Berkovich curves over a complete, algebraically closed field of characteristic , we prove a Riemann-Hurwitz formula relati…