6 papers
Technical Report on Text Dataset Distillation
Keith Ando Ogawa, Bruno Lopes Yamamoto, Lucas Lauton de Alcantara +6
In the vision domain, dataset distillation arises as a technique to condense a large dataset into a smaller synthetic one that exhibits a similar result in the training process. Wh…
The Virtues of Brevity: Avoid Overthinking in Parallel Test-Time Reasoning
Raul Cavalcante Dinardi, Bruno Yamamoto, Anna Helena Reali Costa +1
Reasoning models represent a significant advance in LLM capabilities, particularly for complex reasoning tasks such as mathematics and coding. Previous studies confirm that paralle…
One Period to Rule Them All: Identifying Critical Learning Periods in Deep Networks
Vinicius Yuiti Fukase, Heitor Gama, Barbara Bueno +3
Critical Learning Periods comprehend an important phenomenon involving deep learning, where early epochs play a decisive role in the success of many training recipes, such as data…
Pruning Everything, Everywhere, All at Once
Gustavo Henrique do Nascimento, Ian Pons, Anna Helena Reali Costa +1
Deep learning stands as the modern paradigm for solving cognitive tasks. However, as the problem complexity increases, models grow deeper and computationally prohibitive, hindering…
Improving Fairness in LLMs Through Testing-Time Adversaries
Isabela Pereira Gregio, Ian Pons, Anna Helena Reali Costa +1
Large Language Models (LLMs) push the bound-aries in natural language processing and generative AI, driving progress across various aspects of modern society. Unfortunately, the pe…
Layer Pruning with Consensus: A Triple-Win Solution
Leandro Giusti Mugnaini, Carolina Tavares Duarte, Anna H. Reali Costa +1
Layer pruning offers a promising alternative to standard structured pruning, effectively reducing computational costs, latency, and memory footprint. While notable layer-pruning ap…