ASTRA: A Negotiation Agent with Adaptive and Strategic Reasoning via Tool-integrated Action for Dynamic Offer Optimization
Deuksin Kwon, Jiwon Hae, Emma Clift, Daniel Shamsoddini, Jonathan Gratch, Gale M. Lucas · 2025
Read ASTRA: A Negotiation Agent with Adaptive and Strategic Reasoning via Tool-integrated Action for Dynamic Offer Optimization by Deuksin Kwon, Jiwon Hae, Emma Clift, Daniel Shamsoddini, Jonathan Gratch, Gale M. Lucas (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-2503-07129.txt.
Description
Deuksin Kwon and five co-authors build, in 2025 work, a haggling agent that plans each move as talks unfold. Good haggling must weigh one's own gain against give-and-take, a blend past agents missed: they learnt from flawed human play, bent poorly to a partner's shifts, and planned thin. Their agent runs on two rules: model the partner, and mirror their stance.
Each turn, the agent reads the partner's last move, solves for a sharp counter-offer with a math solver, then picks an offer by the odds it lands. In fake and human tests it bent well to shifting partners and won fair deals. It also works as a coach, the team shows, with plain tips that beat human hunches.
Catalog record
| CREATOR | Deuksin Kwon, Jiwon Hae, Emma Clift, Daniel Shamsoddini, Jonathan Gratch, Gale M. Lucas |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 63 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-07129