6 papers
SE-UNet: Singular Equivariant Imaging for Real-World Constrained Generation
Kanishk Awadhiya
While diffusion models have revolutionized image synthesis, their application to real-world inverse problems is often hampered by the need for massive datasets and the difficulty o…
Reasoning as Attractor Dynamics: Latent Memory Retrieval via Gibbs-Weighted Energy Minimization
Kanishk Awadhiya
Large Language Models (LLMs) are traditionally viewed as autoregressive generators. However, from the perspective of collective computation, they function as high-dimensional Dense…
Parallel Manifold Steering: Efficient Adaptation of Large Associative Memories via Residual Energy Shaping
Kanishk Awadhiya
Large Transformer models function as Dense Associative Memories (DAMs), retrieving knowledge via high-dimensional attractor dynamics driven by the self-attention mechanism \citep{r…
The Fractal Neural Operator: Overcoming Spectral Bias in Chaotic Attractors via Prime-Harmonic Weierstrass Encodings
Kanishk Awadhiya
Deep learning models, particularly Transformers and Neural Operators, exhibit a well-documented "spectral bias," effectively acting as low-pass filters that smooth out high-frequen…
Bifocal Attention: Harmonizing Geometric and Spectral Positional Embeddings for Algorithmic Generalization
Kanishk Awadhiya
Rotary Positional Embeddings (RoPE) have become the standard for Large Language Models (LLMs) due to their ability to encode relative positions through geometric rotation. However,…
The Inductive Bottleneck: Data-Driven Emergence of Representational Sparsity in Vision Transformers
Kanishk Awadhiya
Vision Transformers (ViTs) lack the hierarchical inductive biases inherent to Convolutional Neural Networks (CNNs), theoretically allowing them to maintain high-dimensional represe…