Metraipolis › archive › dc_ca93068a29bd

The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language Models — arXiv:2609.21509

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.21509
License (as archived): CC0-1.0
Archived: 2026-10-10T01:05:23.118Z
SHA-256 of the archived text: ef804264b3a792e38e8c1989cdc1fbaaf269d1f70963b3cdb67e76803de09eab
Stable URL of this page: https://metraipolis.com/documento/dc_ca93068a29bd

How to cite

The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language Models — arXiv:2609.21509 — arXiv (en), https://arxiv.org/abs/2609.21509 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_ca93068a29bd (Metraipolis archive dc_ca93068a29bd, sha256 ef804264b3a792e38e8c1989cdc1fbaaf269d1f70963b3cdb67e76803de09eab)

BibTeX

@misc{dcca93068a29bd,
  title = {The Communication Bottleneck: A Round-Trip Study of Tree-Structured Expression Serialization in Language Models — arXiv:2609.21509},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.21509},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_ca93068a29bd; sha256 ef804264b3a792e38e8c1989cdc1fbaaf269d1f70963b3cdb67e76803de09eab},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.21509. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

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. A generator converts a procedurally generated arithmetic expression into a word problem, a separate extractor recovers the expression from the word problem alone, and symbolic equivalence provides an exact oracle. Evaluating all pairwise combinations of sixteen models yields a communication matrix whose marginals separate generation quality from extraction quality. Three main findings emerge. First, the channel is lossy and asymmetric: swapping which model generates and which extracts shifts accuracy by up to 60.4 points, and the best pair reaches 92.9% by combining different models on each end rather than the same model on both. Second, at least 73.6% of round-trip failures originate at generation, and difficulty is driven by tree structure (operator count, depth, right-branching) rather than model family. Third, the channel is trainable: ~3600 fine-tuning examples that share the evaluation's operators and tree shapes lift every open-weight model above untrained Gemini-3.1-Pro, an upper bound under matched semantics. A disjoint-domain regime with new operators and vocabulary also raises every open-weight model, confirming the gain is not an artifact of matched semantics, though a gap to the frontier remains. Together these results identify tree-structured expression serialization as a primary limiting factor when models communicate hierarchical structure through natural language.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-21 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.21509 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T00:23:42.459Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-21?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).