collaborators

12 papers

cs.LG2026

p-Spin Glass Network Efficient Single-Batch Continual Learning

Vladimer Khasia

Modern sequence models heavily rely on massive memory footprints and large-batch stochastic optimization, barriers that restrict sample efficiency and continual learning. We introd…

cs.LG2026

HoloByte: Continuous Hyperspherical Distillation for Tokenizer-Free Modeling

Vladimer Khasia

Sequence modeling universally relies on discrete subword tokenization to circumvent the computational intractability of native byte-level attention. However, thi…

cs.LG2026

BASIS: Balanced Activation Sketching with Invariant Scalars for "Ghost Backpropagation"

Vladimer Khasia

The activation memory required for exact backpropagation scales linearly with network depth, context length, and feature dimensionality, forming an O(L * BN ) spatial bottleneck (w…

cs.LG2026

Beyond Attention: True Adaptive World Models via Spherical Kernel Operator

Vladimer Khasia

The pursuit of world model based artificial intelligence has predominantly relied on projecting high-dimensional observations into parameterized latent spaces, wherein transition d…

cs.LG2026

Hybrid Dual-Path Linear Transformations for Efficient Transformer Architectures

Vladimer Khasia

Standard Transformer architectures rely heavily on dense linear transformations, treating feature projection as a monolithic, full-rank operation. We argue that this formulation is…

cs.LG2026

HAS-VQ: Hessian-Adaptive Sparse Vector Quantization for High-Fidelity LLM Compression

Vladimer Khasia

Post-training quantization is essential for deploying Large Language Models (LLMs) on resource-constrained devices. However, standard integer quantization (e.g., INT4) fundamentall…