4 papers
Tensorizing Engram: Sharing Latents Across N-Gram Embeddings is Beneficial in LLMs
Wuyang Zhou, Yuxuan Gu, Giorgos Iacovides +3
Modern language models represent text using discrete token-level embeddings, which forces recurring multi-token patterns to be learned implicitly across Transformer layers. Both Ov…
Calibrating Uncertainty for Zero-Shot Adversarial CLIP
Wenjing Lu, Zerui Tao, Yuning Qiu +3
CLIP delivers strong zero-shot classification but remains highly vulnerable to adversarial attacks. Prior adversarial fine-tuning work primarily matches predicted logits between cl…
Spectral Anatomy of Quantum Gaussian Process Kernels
Jian Xu, Chao Li, Guang Lin +4
Two recent results have reshaped quantum Gaussian processes (QGPs). On the one hand, \citet{lowe2025assessing} rule out the exponential speedups claimed by HHL-based QGP regression…
Efficient Low Rank Attention for Long-Context Inference in Large Language Models
Tenghui Li, Guoxu Zhou, Xuyang Zhao +2
As the length of input text increases, the key-value (KV) cache in LLMs imposes prohibitive GPU memory costs and limits long-context inference on resource constrained devices. Exis…