Ning Liu

This paper demonstrates a novel attack against the shuffling defense used in secure Transformer inference, showing that randomly permuted activations can still be exploited to recover model weights.

DREAM-R is a novel framework that significantly enhances speculative reasoning in large multimodal models by optimizing draft generation alignment, introducing a robust verification mechanism, and enabling fully parallel execution.

This paper proposes RECONTEXT, a training-free inference method for improving long-context reasoning in large language models using model-internal relevance signals and recursive evidence replay.

The paper presents a tool-making pipeline for production LLM agents that compiles repeated steps into validated, versioned tools before deployment, reducing latency and error rate.