Challenge of Making Agents

Agents are not very new—(Riedl and Amant 2002). But newer models can be powered by LLM/VLMs, meaning we are using language for reasoning/communication.

Sequentiality is hard

1. what is the context/motivation?
2. how to you transfer across contexts?
3. how do you plan?

Evaluation

1. Different from how previous NLP benchmarks: we are not worried about language modeling
2. No longer boundaries between various fields

Common goals:

• realistic agents—stop playing Atari games.
• reproducible systems
• measurability goals
• scalable models
• which are easy to use

Web as an Interactive Environment

• agents on the web is both practical and scalable
• https://webshop-pnlp.github.io/
• WebShop can actually transfer with no work to training on Amazon
• Mind2Web

InterCode

Formulation of agent decisions as POMDP in order to fully benchmark Markovian decisions:

• 80% of the human genome is actually transcribed
• very little “junk DNA”
• 40% IncRNA are gene specific

char is a character that represents a glypth:

The polynomial given by the determinant of:

\begin{equation} det(A-\lambda I) \end{equation}

for some Linear Map \(A\). Solutions for \(\lambda\) are the eigenvalues. This is because something is an eigenvalue IFF \((A-\lambda I)v = 0\) for some \(\lambda, v\), so we need \((A-\lambda I)\) to be singular.

Characteristic polynomial of a 2x2 matrix is given by \(\lambda^{2}-tr(A)\lambda + det(A)\).