Výsledky vyhledávání

Language model alignment has become a critical step in training modern generative language models. The goal of alignment is to finetune a reference model such that the win rate of a sample from the aligned model over a sample from the reference model

Externí odkaz: http://arxiv.org/abs/2412.19792

Zobrazit plný text záznamu

Report

HashEvict: A Pre-Attention KV Cache Eviction Strategy using Locality-Sensitive Hashing

Autor: Liu, Minghui, Rabbani, Tahseen, O'Halloran, Tony, Sankaralingam, Ananth, Hartley, Mary-Anne, Gravelle, Brian, Huang, Furong, Fermüller, Cornelia, Aloimonos, Yiannis

Transformer-based large language models (LLMs) use the key-value (KV) cache to significantly accelerate inference by storing the key and value embeddings of past tokens. However, this cache consumes significant GPU memory. In this work, we introduce

Externí odkaz: http://arxiv.org/abs/2412.16187

Zobrazit plný text záznamu

Report

FeynKrack: A continuum model for quasi-brittle damage through Feynman-Kac killed diffusion

Autor: Prakash, Ved, Gopal, Upadhyayula M. M. A. Sai, Das, Sanhita, Ramaswamy, Ananth, Roy, Debasish

Continuum damage mechanics (CDM) is a popular framework for modelling crack propagation in solids. The CDM uses a damage parameter to quantitatively assess what one loosely calls `material degradation'. While this parameter is sometimes given a physi

Externí odkaz: http://arxiv.org/abs/2412.00791

Zobrazit plný text záznamu

Report

Exponentially Consistent Nonparametric Clustering of Data Streams

Autor: Singh, Bhupender, Rajagopalan, Ananth Ram, Bhashyam, Srikrishna

In this paper, we consider nonparametric clustering of $M$ independent and identically distributed (i.i.d.) data streams generated from unknown distributions. The distributions of the $M$ data streams belong to $K$ underlying distribution clusters. E

Externí odkaz: http://arxiv.org/abs/2411.13922

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání