Learn Before
Inarticulacy of Human Preferences as an Alignment Challenge
LLM alignment is complicated by the fact that humans often struggle to precisely or consistently articulate their preferences beforehand. This inherent ambiguity makes it difficult to define a clear objective for the model. Often, desired behavior only becomes clear to a user after they have observed an undesirable or unexpected response from the LLM, highlighting the challenge of creating comprehensive alignment guidelines in advance.
0
1
Tags
Ch.4 Alignment - Foundations of Large Language Models
Foundations of Large Language Models
Foundations of Large Language Models Course
Computing Sciences
Ch.2 Generative Models - Foundations of Large Language Models
Related
Shift in LLM Alignment from Predefined Tasks to Real-World Interaction
Impracticality of Achieving Alignment Solely Through Pre-training
Need for Diverse Alignment Methods
Insufficiency of Data Fitting for Value Alignment
Difficulty of Encoding Human Values in Datasets
Inarticulacy of Human Preferences as an Alignment Challenge
Goodhart's Law
Real-World Complexity as an Alignment Challenge
Specification Gaming in AI Alignment
Alignment Challenges as a Motivator for AI Research
Diversity and Fluidity of Human Values as an Alignment Challenge
Analysis of an LLM Alignment Failure
A development team building a chatbot aims for it to be 'helpful' to all users. They discover that behaviors praised as helpful by users in one country are criticized as intrusive by users in another. This issue persists even after training the model on vast, culturally diverse datasets. Which fundamental challenge in guiding a model's behavior does this scenario best illustrate?
Evaluating Core Difficulties in Model Behavior Guidance
Challenge of Defining Human Values for AI Objectives