When control meets large language models: From words to dynamics
Komeil Nosrati, Aleksei Tepljakov, Juri Belikov, Eduard Petlenkov · 2026
Read When control meets large language models: From words to dynamics by Komeil Nosrati, Aleksei Tepljakov, Juri Belikov, Eduard Petlenkov (2026) 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-2602-03433.txt.
Description
Control engineers design systems that steer machines, while large language models now need steering of their own. Komeil Nosrati and 3 co-authors map this two-way relationship in a 2026 paper. Prompts can help design controllers, and control theory can in turn shape prompts and model behavior toward specific goals.
The authors first review how language models assist control directly, by drafting and tuning controllers, and indirectly, by speeding up research work. They then show how control ideas steer model trajectories through activation-level interventions, parameter editing, and input optimization. A final section treats the model itself as a dynamical system inside a control loop. The paper closes with open challenges for trustworthy, controllable models.
Catalog record
| CREATOR | Komeil Nosrati, Aleksei Tepljakov, Juri Belikov, Eduard Petlenkov |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 114 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2602-03433