Výsledky vyhledávání - "Kim, Junseok"

Report

Stage-Wise Reward Shaping for Acrobatic Robots: A Constrained Multi-Objective Reinforcement Learning Approach

Autor: Kim, Dohyeong, Kwon, Hyeokjin, Kim, Junseok, Lee, Gunmin, Oh, Songhwai

As the complexity of tasks addressed through reinforcement learning (RL) increases, the definition of reward functions also has become highly complicated. We introduce an RL method aimed at simplifying the reward-shaping process through intuitive str

Externí odkaz: http://arxiv.org/abs/2409.15755

Zobrazit plný text záznamu

Report

Persona is a Double-edged Sword: Mitigating the Negative Impact of Role-playing Prompts in Zero-shot Reasoning Tasks

Autor: Kim, Junseok, Yang, Nakyeong, Jung, Kyomin

Recent studies demonstrate that prompting a role-playing persona to an LLM improves reasoning capability. However, assigning an adequate persona is difficult since LLMs are extremely sensitive to assigned prompts; thus, inaccurately defined personas

Externí odkaz: http://arxiv.org/abs/2408.08631

Zobrazit plný text záznamu

Report

Acylindrical hyperbolicity of outer automorphism groups of right-angled Artin groups

Autor: Baik, Hyungryul, Kim, Junseok

We study the acylindrical hyperbolicity of the outer automorphism group of a right-angled Artin group $A_\Gamma$. When the defining graph $\Gamma$ has no SIL-pair (separating intersection of links), we obtain a necessary and sufficient condition for

Externí odkaz: http://arxiv.org/abs/2405.19702

Zobrazit plný text záznamu

Report

Skeleton: A New Framework for Accelerating Language Models via Task Neuron Localized Prompt Tuning

Autor: Yang, Nakyeong, Moon, Jiwon, Kim, Junseok, Jang, Yunah, Jung, Kyomin

Prompt tuning methods have shown comparable performance to general training methods as parameter-efficient fine-tuning (PEFT) methods in various natural language understanding tasks. However, existing prompt tuning methods still utilize the entire mo

Externí odkaz: http://arxiv.org/abs/2404.11916

Zobrazit plný text záznamu