2025/01/23 by Vignav Ramesh, Kenneth Li, Ramesh, Vignav +1 · 3 citations
Computer Science · #Artificial Intelligence (cs.AI) #Computation and Language (cs.CL) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Multi-Agent Systems and Negotiation #Natural Language Processing Techniques #Semantic Web and Ontologies
paper · pdf · doi:10.48550/arxiv.2501.14082
openalex publication_date 2025/01/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Communication between multiple language model (LM) agents has been shown to scale up the reasoning ability of LMs. While natural language has been the dominant medium for inter-LM communication, it is not obvious this should be the standard: not only does natural language communication incur high inference costs that scale quickly with the number of both agents and messages, but also the decoding process abstracts away too much rich information that could be otherwise accessed from the internal activations. In this work, we propose a simple technique whereby LMs communicate via activations; concretely, we pause an LM B's computation at an intermediate layer, combine its current activation with another LM A's intermediate activation via some function f, then pass f's output into the next layer of B and continue the forward pass till decoding is complete. This approach scales up LMs on new tasks with zero additional parameters and data, and saves a substantial amount of compute over natural language communication. We test our method with various functional forms f on two experimental setups--multi-player coordination games and reasoning benchmarks--and find that it achieves up to 27.0% improvement over natural language communication across datasets with <1/4 the compute, illustrating the superiority and robustness of activations as an alternative "language" for communication between LMs.