LLM agents are increasingly deployed in collaborative settings where they pursue separate tasks while sharing resources and operating under review or oversight processes. Such settings create safety risks not only when agents are malicious, but also when they are simply trying to help. Prior work has largely studied ad...
Moumita Choudhury, Amit Sarker, S. Zilberstein· 0 citations
This work introduces a repeated, multi-party, multi-issue negotiation framework in which recurring agents participate in a sequence of negotiation games while one participant covertly attempts to prevent agreement.
Amit Sarker, Moumita Choudhury, Saaduddin Mahmud et al.· 0 citations
It is shown that persuasive interaction leaves a measurable, task-dependent behavioral footprint on downstream agent execution, and agents that are more susceptible to persuasive conversational context do not reliably predict the magnitude of the downstream behavioral shifts.
Hyejun Jeong, Amir Houmansadr, S. Zilberstein et al.· 0 citations
RoboTalk is introduced, a synthetic data-generation pipeline and dataset of 7,950 multimodal trajectories spanning 53 mobile-manipulation kitchen tasks for training small VLMs to communicate and coordinate and fine-tuning open-source models on this dataset can reach 77% success on novel held-out tasks.
Dorian Benhamou Goldfajn, Mason Nakamura, Saaduddin Mahmud et al.· 0 citations
CALM (Controller-Aware Language Models), a post-training framework that explicitly places controllers in the training loop, is introduced, showing that controller-aware post-training improves generalization across inference-time workflows beyond single-controller optimization.
Moumita Choudhury, Vanshaj Khattar, Jing Liu et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
