Search Results for author: Junseok Kim

Found 4 papers, 1 papers with code

Self-Training using Rules of Grammar for Few-Shot NLU

no code implementations Findings (EMNLP) 2021 Joonghyuk Hahn, Hyunjoon Cheon, Kyuyeol Han, Cheongjae Lee, Junseok Kim, Yo-Sub Han

We propose to use rules of grammar in self-training as a more reliable pseudo-labeling mechanism, especially when there are few labeled data.

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

1 code implementation24 Sep 2024 Dohyeong Kim, Hyeokjin Kwon, Junseok Kim, Gunmin Lee, Songhwai Oh

As the complexity of tasks addressed through reinforcement learning (RL) increases, the definition of reward functions also has become highly complicated.

Multi-Objective Reinforcement Learning

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

no code implementations16 Aug 2024 Junseok Kim, Nakyeong Yang, Kyomin Jung

Then, Jekyll \& Hyde collects two potential solutions from role-playing and neutral prompts and selects a better solution using the LLM evaluator.

Position

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

no code implementations18 Apr 2024 Nakyeong Yang, Jiwon Moon, Junseok Kim, Yunah Jang, Kyomin Jung

Prompt tuning methods have shown comparable performance to general training methods as parameter-efficient fine-tuning (PEFT) methods in various natural language understanding tasks.

Language Modeling Language Modelling +2

Cannot find the paper you are looking for? You can Submit a new open access paper.