Multi-Scale Model Compression via Nested Matrix Learning

UvA-DARE (Digital Academic

LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches

Reinforcement learning has greatly advanced the capabilities of large language models, but its memory demands remain a barrier to broader adoption. We introduce LoGRA, an approach to RL post-training that reduces memory by retaining useful learning signals in low-rank gradient sketches. These compact representations su...

GradSup: Gradient Superposition for Personalised and Scalable LLM Recommendation

Large language models (LLMs) have demonstrated strong capabilities in recommendation tasks such as item, sequence, conversational recommendation, and explanation generation. However, LLM weights are typically shared across all users. Adapting these models to individual users remains a fundamental challenge that require...

MoR-MLLM: Mixture of Recursions for Efficient Multimodal Large Language Models

This work proposes MoR-MLLM, a computation-sparse MLLM based on the recent Mixture-of-Recursions (MoR) framework, which introduces adaptive per-token recursion, allowing the model to dynamically adjust its recursive depth and allocate more computation to visually or linguistically challenging tokens while skipping redu...

BatchParScale: Sparse Batch-Mixing for Sequence Models

Recent test-time scaling methods improve model performance by increasing inference-time compute. One practical form of this is parallel scaling, where a model samples multiple independent reasoning streams and aggregates their answers using methods like voting, reranking, or verification. While effective, this approach...

Infinite-Parameter LLMs: Generating and Adapting Weights from Live Data

This work proposes the \textbf{Infinite-Parameter LLM}.