VQEL: Enabling Self-Play in Emergent Language Games via Agent-Internal Vector Quantization
Mahdi Samiei [email protected], Mehdi Jamalkhah [email protected], Mahdieh Soleymani Baghshah [email protected] · 2025
Read VQEL: Enabling Self-Play in Emergent Language Games via Agent-Internal Vector Quantization by Mahdi Samiei [email protected], Mehdi Jamalkhah [email protected], Mahdieh Soleymani Baghshah [email protected] (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-2503-04940.txt.
Description
Before babies talk to others, they babble to themselves, and Mohammad Mahdi Samiei Paqaleh, Mehdi Jamalkhah, and Mahdieh Soleymani Baghshah give artificial agents the same head start. Teaching agents a discrete symbol language is notoriously unstable, since sampling symbols blocks the usual training signals. Their architecture routes messages through a learned codebook, snapping internal signals to the nearest entry while keeping training smooth.
Agents first play alone, building a private vocabulary from the codebook. Paired up afterwards, they aligned symbols more consistently and solved their shared tasks more often than agents trained only together. The authors present self-play as a principled foundation for emergent communication, borrowing the developmental order that works for humans.
Catalog record
| CREATOR | Mahdi Samiei [email protected], Mehdi Jamalkhah [email protected], Mahdieh Soleymani Baghshah [email protected] |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 58 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-04940