4 papers
Non-Interfering Weight Fields: Treating Model Parameters as a Continuously Extensible Function
Sarim Chaudhry
Large language models store all learned knowledge in a single, fixed weight vector. Teaching a model new capabilities requires modifying those same weights, inevitably degrading pr…
Recursive Concept Evolution for Compositional Reasoning in Large Language Models
Sarim Chaudhry
Large language models achieve strong performance on many complex reasoning tasks, yet their accuracy degrades sharply on benchmarks that require compositional reasoning, including…
Semantic Belief-State World Model for 3D Human Motion Prediction
Sarim Chaudhry
Human motion prediction has traditionally been framed as a sequence regression problem where models extrapolate future joint coordinates from observed pose histories. While effecti…
ChronoPlastic Spiking Neural Networks
Sarim Chaudhry
Spiking neural networks (SNNs) offer a biologically grounded and energy-efficient alternative to conventional neural architectures; however, they struggle with long-range temporal…