Latent Communication Between Language Model Agents: Channels, Alignment, and the Limits of Text
Quick Answer
This study reveals that Large Language Models (LLMs) lose significant information when communicating via text, with a 99.4% probe accuracy in sparse channels compared to 80.4% in text channels.
Quick Take
This study reveals that Large Language Models (LLMs) lose significant information when communicating via text, with a 99.4% probe accuracy in sparse channels compared to 80.4% in text channels. The analysis indicates that text serialization destroys 88% of important features, suggesting that latent communication may require deeper tasks for practical advantages.
Key Points
- Sparse Autoencoder channels retain 99.4% accuracy at 28-fold compression over dense-latent channels.
- Text serialization destroys 88% of SAE features, replacing them with irrelevant features.
- Linear Procrustes alignment incurs a 3-10pp performance penalty compared to nonlinear methods.
- Latent channels match text channels on cross-lingual tasks but do not exceed them.
- Deeper tasks are necessary to explore the practical advantages of latent communication.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Multi-agent systems (MAS) are utilized in many contexts and many professions. Those MAS rely on inter-agent communication, usually implemented by clear-text message passing. We hypothesize that Large Language Models may have a world model at their disposal that exceeds expressibility in text when complex concepts need to be communicated. Our aim is to approach a proof of this hypothesis with structured experiments. In this work, we show that LLM agents communicating via text lose information, which we quantify via Sparse Autoencoder (SAE) feature analysis. We construct three communication channels and measure concept-discriminating information in each. We first show that the SAE-sparse channel retains a 99.4% probe accuracy at 28-fold compression over the dense-latent channel vs 80.4% for the text channel. We then proceed to examine the same for cross-architecture communication by using sparse latent space alignment. We find for Procrustes alignment a 92% top-1 retrieval between Llama and Mistral. Using a text round-trip, we perform feature survival analysis to find that text serialization destroys 88% of SAE features, replacing them with a different feature set. We attribute the loss to identity replacement, not attenuation. By our analysis, we were able to attribute a 3-10pp performance penalty to the linear Procrustes alignment, improving with nonlinear alignment methods. In a task-level evaluation we find that the latent channel matches the text channel on cross-lingual concept tasks but never exceeds it. Text augmentation with latent features provides no benefit, leading us to negative conclusions for the initial hypothesis: lost features mostly or completely encode surface form, not task-relevant semantics. To pinpoint the practical advantage of latent communication over a text channel, deeper tasks eliciting complex concepts and an corresponding analysis framework are needed.
| Subjects: | Computation and Language (cs.CL); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2607.14103 [cs.CL] |
| (or arXiv:2607.14103v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2607.14103 arXiv-issued DOI via DataCite |
Submission history
From: Markus Wenzel [view email]
[v1]
Wed, 6 May 2026 13:12:00 UTC (1,547 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CL
See more →Letting the Data Speak: Extracting Keywords from Crowdsourced Collections with AI
The study evaluates three NLP approaches—Named Entity Recognition, Keyword Extraction, and Topic Modelling—using the Their Finest Hour Online Archive to automate keyword extraction from crowdsourced WWII collections. Findings suggest that while NLP methods show promise, no single approach is sufficient, and ethical considerations in automated keyword extraction are crucial for responsible stewardship.