Theory of Mind in Action: The Instruction Inference Task Journal: Artificial Intelligence
Fardin Saad, Pradeep K. Murukannaiah, Munindar P. Singh · 2025
Read Theory of Mind in Action: The Instruction Inference Task Journal: Artificial Intelligence by Fardin Saad, Pradeep K. Murukannaiah, Munindar P. Singh (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2507-02935.txt.
Description
Guessing what a person means by loose orders is tested by Fardin Saad, Pradeep K. Murukannaiah, and Munindar P. Singh in a 2025 arXiv paper. Good teaming with a software agent needs it to grasp orders that leave gaps. The agent must infer the aims the person left unsaid. The authors call that work an Instruction Inference task and set it in a live goal task.
Their agent, Tomcat, comes in two forms. One learns from a few worked cases with step-by-step thought. One leans on plain common sense. Both ran on three big language models. Scores for aim match and plan quality put the first form near human levels in a study with fifty-two people.
Catalog record
| CREATOR | Fardin Saad, Pradeep K. Murukannaiah, Munindar P. Singh |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 104 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2507-02935