The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language Models
The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language Models

Key points
- When language models reason in chain-of-thought or exchange free-text intermediates, they serialize structured information into natural language.
- How much tree-structured compositional content survives this bottleneck?
- We propose a round-trip protocol that answers this question empirically for tree-structured expressions.
- Together these results identify tree-structured expression serialization as a primary limiting factor when models communicate hierarchical structure through natural language.
Sources (1)
- [1]The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language ModelsApple Machine Learning Research · Sep 29, 12:00 AM
The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language Models
When language models reason in chain-of-thought or exchange free-text intermediates, they serialize structured information into natural language.
Extractive summary: sentences quoted from the sources.
Before this
- Sep 28, 2026openai/openai-python v3.20.0
- Sep 24, 2026openai/openai-python v3.19.2
- Sep 2, 2026pytorch/pytorch v2.14.0: PyTorch 2.14.0 Release
- Jul 15, 2026huggingface/transformers v5.14.0: Release v5.14.0
- Jul 8, 2026pytorch/pytorch v2.13.0: PyTorch 2.13.0 Release
- Jun 10, 2026DiffusionGemma: 4x faster text generation