Add to ORKGLarge Language Model (LLM) has demonstrated significant ability in various Natural Language Processing tasks. However, their effectiveness is highly dependent on the phrasing of the task prompt, leading to research on automatic prompt optimization using labeled task data. We reveal that these prompt optimization techniques are vulnerable to distribution shifts such as subpopulation shifts, which are common for LLMs in real-world scenarios such as customer reviews analysis. In this light, we propose a new problem of robust prompt optimization for LLMs against distribution shifts, which requires the prompt optimized over the labeled source group can simultaneously generalize to an unlabeled target group. To solve this problem, we propose Gene...
Open-sourced large language models (LLMs) have demonstrated remarkable efficacy in various tasks wit...
Natural language prompts have been shown to facilitate cross-task generalization for large language ...
Adopting a two-stage paradigm of pretraining followed by fine-tuning, Pretrained Language Models (PL...
The advancement of neural network models has led to state-of-the-art performance in a wide range of ...
When primed with only a handful of training samples, very large, pretrained language models such as ...
Prompt tuning learns soft prompts to condition frozen Pre-trained Language Models (PLMs) for perform...
Why can pre-trained language models (PLMs) learn universal representations and effectively adapt to ...
This paper presents AutoHint, a novel framework for automatic prompt engineering and optimization fo...
Large language models (LLMs), while transformative for NLP, come with significant computational dema...
Reinforcement learning (RL) has emerged as a powerful paradigm for fine-tuning Large Language Models...
Prompting has shown impressive success in enabling large pretrained language models (LMs) to perform...
Pretrained large language models (LLMs) are strong in-context learners that are able to perform few-...
Recent works have shown promising results of prompt tuning in stimulating pre-trained language model...
We study the impact of source length and verbosity of the tuning dataset on the per-formance of para...
Recent studies show that prompt tuning can better leverage the power of large language models than f...
Open-sourced large language models (LLMs) have demonstrated remarkable efficacy in various tasks wit...
Natural language prompts have been shown to facilitate cross-task generalization for large language ...
Adopting a two-stage paradigm of pretraining followed by fine-tuning, Pretrained Language Models (PL...
The advancement of neural network models has led to state-of-the-art performance in a wide range of ...
When primed with only a handful of training samples, very large, pretrained language models such as ...
Prompt tuning learns soft prompts to condition frozen Pre-trained Language Models (PLMs) for perform...
Why can pre-trained language models (PLMs) learn universal representations and effectively adapt to ...
This paper presents AutoHint, a novel framework for automatic prompt engineering and optimization fo...
Large language models (LLMs), while transformative for NLP, come with significant computational dema...
Reinforcement learning (RL) has emerged as a powerful paradigm for fine-tuning Large Language Models...
Prompting has shown impressive success in enabling large pretrained language models (LMs) to perform...
Pretrained large language models (LLMs) are strong in-context learners that are able to perform few-...
Recent works have shown promising results of prompt tuning in stimulating pre-trained language model...
We study the impact of source length and verbosity of the tuning dataset on the per-formance of para...
Recent studies show that prompt tuning can better leverage the power of large language models than f...
Open-sourced large language models (LLMs) have demonstrated remarkable efficacy in various tasks wit...
Natural language prompts have been shown to facilitate cross-task generalization for large language ...
Adopting a two-stage paradigm of pretraining followed by fine-tuning, Pretrained Language Models (PL...