Dhruv Mali

TTDnet — ThingsThoughtDiffusion Reconstructing Visual Perception from EEG Signals via Multi-Scale Neural Encoding and Latent Diffusion Overview TTDnet is a two-stage framework for generating photorealistic images directly from electroencephalography (EEG) recordings. Given a raw multi-channel EEG epoch captured while a subject views an object, TTDnet reconstructs a semantically faithful image of that object. The system bridges the gap between low-SNR neural signals and high-fidelity image generation by c...

A hybrid CNN-Transformer architecture for multi-class EEG-based stress classification using multi-scale temporal feature extraction, squeeze-and-excitation attention, and self-attention mechanisms. Overview InceptSADNet extends the conventional single-scale temporal convolution paradigm used in EEG classification models (such as EEGNet and ShallowConvNet) by introducing an Inception-inspired multi-scale temporal front-end coupled with a Transformer encoder for long-range temporal dependency modeling. The...

Introduction From a very long-time new AI tools like ChatGPT, cursor, Replit etc are lifting the hood and coming up with new vibe coding ideas and making their software more and more powerful, while google just raised the bar high and introduced us to AlphaEvolve. Versatility of LLMs is crazy they can generate code, brainstorm ideas and even debug or correct us sometimes, well AlphaEvolve is an evolutionary coding agent which not only does work of other coding agents but on top of that agent powered by l...