paper-with-me

Papers

Understanding Prompt Tuning and In-Context Learning via Meta-Learning

2025-05-22 · Tim Genewein, Kevin Wenliang Li, Jordi Grau-Moya, Anian Ruoss, Laurent Orseau, Marcus Hutter

Prompting is one of the main ways to adapt a pretrained model to target tasks. Besides manually constructing prompts, many prompt optimization methods have been proposed in the literature. Method development is mainly empirically driven, with less emphasis on a conceptual understanding of prompting. In this paper we discuss how optimal prompting can be understood through a Bayesian view, which also implies some fundamental limitations of prompting that can only be overcome by tuning weights. The paper explains in detail how meta-trained neural networks behave as Bayesian predictors over the pretraining distribution, whose hallmark feature is rapid in-context adaptation. Optimal prompting can be studied formally as conditioning these Bayesian predictors, yielding criteria for target tasks where optimal prompting is and is not possible. We support the theory with educational experiments on LSTMs and Transformers, where we compare different versions of prefix-tuning and different weight-tuning methods. We also confirm that soft prefixes, which are sequences of real-valued vectors outside the token alphabet, can lead to very effective prompts for trained and even untrained networks by manipulating activations in ways that are not achievable by hard tokens. This adds an important mechanistic aspect beyond the conceptual Bayesian theory.

📄 PDF Abstract BibTeX arXiv:2505.17010

Code (1)

google-deepmind/thunnini 공식 구현 jax

Tasks

In-Context LearningMeta-Learning

Similar Papers 제목 키워드 기반

Contextual Transformer for Offline Meta Reinforcement Learning

2022-11-15 · Runji Lin, Ye Li, Xidong Feng, Zhaowei Zhang 외

The pretrain-finetuning paradigm in large-scale sequence models has made significant progress in natural language processing and computer vision tasks. However, such a paradigm is still hindered by several challenges in …

D4RLMeta Reinforcement LearningMuJoCoOffline RL+5

Prompt Tuning with Soft Context Sharing for Vision-Language Models

2022-08-29 · Kun Ding, Ying Wang, Pengzhang Liu, Qiang Yu 외

Vision-language models have recently shown great potential on many tasks in computer vision. Meanwhile, prior work demonstrates prompt tuning designed for vision-language models could acquire superior performance on few-…

Few-Shot LearningMulti-Task Learning

MetaTool: Facilitating Large Language Models to Master Tools with Meta-task Augmentation

2024-07-15 · Xiaohan Wang, Dian Li, Yilin Zhao, Sinbadliu 외

Utilizing tools with Large Language Models (LLMs) is essential for grounding AI agents in real-world applications. The prevailing approach involves few-shot prompting with demonstrations or fine-tuning with expert annota…

Data Augmentation

Efficient In-Context Medical Segmentation with Meta-driven Visual Prompt Selection

2024-07-15 · Chenwei Wu, David Restrepo, Zitao Shuai, Zhongming Liu 외

In-context learning (ICL) with Large Vision Models (LVMs) presents a promising avenue in medical image segmentation by reducing the reliance on extensive labeling. However, the ICL performance of LVMs highly depends on t…

Image SegmentationIn-Context LearningMedical Image SegmentationMeta-Learning+1

Tuning LLM-based Code Optimization via Meta-Prompting: An Industrial Perspective

2025-08-02 · Jingzhi Gong, Rafail Giavrimis, Paul Brookes, Vardan Voskanyan 외 arxiv

There is a growing interest in leveraging multiple large language models (LLMs) for automated code optimization. However, industrial platforms deploying multiple LLMs face a critical challenge: prompts optimized for one …

Prompt Engineering