Research

Paper

AI LLM March 20, 2026

The $\mathbf{Y}$-Combinator for LLMs: Solving Long-Context Rot with $λ$-Calculus

Authors

Amartya Roy, Rasul Tutunov, Xiaotong Ji, Matthieu Zimmer, Haitham Bou-Ammar

Abstract

LLMs are increasingly used as general-purpose reasoners, but long inputs remain bottlenecked by a fixed context window. Recursive Language Models (RLMs) address this by externalising the prompt and recursively solving subproblems. Yet existing RLMs depend on an open-ended read-eval-print loop (REPL) in which the model generates arbitrary control code, making execution difficult to verify, predict, and analyse. We introduce $λ$-RLM, a framework for long-context reasoning that replaces free-form recursive code generation with a typed functional runtime grounded in $λ$-calculus. It executes a compact library of pre-verified combinators and uses neural inference only on bounded leaf subproblems, turning recursive reasoning into a structured functional program with explicit control flow. We show that $λ$-RLM admits formal guarantees absent from standard RLMs, including termination, closed-form cost bounds, controlled accuracy scaling with recursion depth, and an optimal partition rule under a simple cost model. Empirically, across four long-context reasoning tasks and nine base models, $λ$-RLM outperforms standard RLM in 29 of 36 model-task comparisons, improves average accuracy by up to +21.9 points across model tiers, and reduces latency by up to 4.1x. These results show that typed symbolic control yields a more reliable and efficient foundation for long-context reasoning than open-ended recursive code generation. The complete implementation of $λ$-RLM, is open-sourced for the community at: https://github.com/lambda-calculus-LLM/lambda-RLM.

Metadata

arXiv ID: 2603.20105
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-03-20
Fetched: 2026-03-23 16:54

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.20105v1</id>\n    <title>The $\\mathbf{Y}$-Combinator for LLMs: Solving Long-Context Rot with $λ$-Calculus</title>\n    <updated>2026-03-20T16:29:51Z</updated>\n    <link href='https://arxiv.org/abs/2603.20105v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.20105v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>LLMs are increasingly used as general-purpose reasoners, but long inputs remain bottlenecked by a fixed context window. Recursive Language Models (RLMs) address this by externalising the prompt and recursively solving subproblems. Yet existing RLMs depend on an open-ended read-eval-print loop (REPL) in which the model generates arbitrary control code, making execution difficult to verify, predict, and analyse.\n  We introduce $λ$-RLM, a framework for long-context reasoning that replaces free-form recursive code generation with a typed functional runtime grounded in $λ$-calculus. It executes a compact library of pre-verified combinators and uses neural inference only on bounded leaf subproblems, turning recursive reasoning into a structured functional program with explicit control flow. We show that $λ$-RLM admits formal guarantees absent from standard RLMs, including termination, closed-form cost bounds, controlled accuracy scaling with recursion depth, and an optimal partition rule under a simple cost model. Empirically, across four long-context reasoning tasks and nine base models, $λ$-RLM outperforms standard RLM in 29 of 36 model-task comparisons, improves average accuracy by up to +21.9 points across model tiers, and reduces latency by up to 4.1x. These results show that typed symbolic control yields a more reliable and efficient foundation for long-context reasoning than open-ended recursive code generation. The complete implementation of $λ$-RLM, is open-sourced for the community at: https://github.com/lambda-calculus-LLM/lambda-RLM.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-20T16:29:51Z</published>\n    <arxiv:primary_category term='cs.LG'/>\n    <author>\n      <name>Amartya Roy</name>\n    </author>\n    <author>\n      <name>Rasul Tutunov</name>\n    </author>\n    <author>\n      <name>Xiaotong Ji</name>\n    </author>\n    <author>\n      <name>Matthieu Zimmer</name>\n    </author>\n    <author>\n      <name>Haitham Bou-Ammar</name>\n    </author>\n  </entry>"
}