When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning

• URL: http://arxiv.org/abs/2606.29354v1
• Date: Sun, 28 Jun 2026 12:02:42 GMT
• Title: When LLMs Develop Languages: Symbolic Communication for Efficient Multi-Agent Reasoning
• Abstract summary: Chain-of-Thought (CoT) improves large language models (LLMs) on difficult reasoning tasks.<n>We propose Communicative Language Symbolism Routing (R)<n>R treats each Language Symbolism Framework as a reusable symbolic protocol with compact symbols, usage rules, and a message-passing contract.<n>R reduces latency-oriented generated token completion by $3sim 6times$ compared to standard CoT.
• Score: 85.36421257648294
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Chain-of-Thought (CoT) improves large language models (LLMs) on difficult reasoning tasks, but it often incurs long natural-language rationales that are poorly aligned with efficient machine reasoning. We propose Communicative Language Symbolism Routing (CLSR), a test-time framework in which multiple LLM agents autonomously invent, evolve, and share compact Language Symbolism Frameworks (LSFs), while a latent-free router adaptively selects and composes these languages per query to optimize the accuracy-token trade-off. Unlike prompt optimization that refines surface instructions, CLSR treats each LSF as a reusable symbolic protocol with compact symbols, usage rules, and a message-passing contract, and improves it through an evolutionary loop driven by correctness and token cost. At inference time, the router may invoke a single low-cost LSF call, ensemble multiple LSFs, or execute a multi-round LSF composition protocol on harder queries. Across challenging benchmarks, CLSR reduces latency-oriented generated token completion by $3\sim 6\times$ compared to standard CoT while maintaining accuracy. We further derive an information-theoretic lower bound on token cost under arbitrary symbolism and show that, under an interpreter-realizability premise, multi-round LSF protocols conditionally subsume program-execution pipelines. Code is publicly available (https://github.com/pzqpzq/LSF_MDia).

Related papers

• The Imitation Game: When LLMs Learn to Reason Like Programs via Code-Centric Reasoning Data Synthesis [50.473217152006875]
  MIMIC is a framework that leverages executable code as a rigorous medium for reasoning data synthesis.<n>MIMIC transforms algorithms into verifiable reasoning trajectories through narrative fusion, code-guided test synthesis, and dynamic code instrumentation.<n>Our method significantly elevates accuracy across general reasoning, complex mathematical benchmarks, and fine-grained deterministic tasks.
  arXiv  Detail & Related papers  (2026-09-13T17:04:16Z)
• RuC: HDL-Agnostic Rule Completion Benchmark Generation [1.8927558216457558]
  Large Language Models (LLMs) have rapidly improved in performance across code-related tasks.<n>We present a framework for language-agnostic rule completion (RuC)<n>RuC produces RTL code-completion tasks from a set of input hardware description sources.
  arXiv  Detail & Related papers  (2026-04-30T12:19:40Z)
• Less Languages, Less Tokens: An Efficient Unified Logic Cross-lingual Chain-of-Thought Reasoning Framework [78.44489936691467]
  Cross-lingual chain-of-thought (XCoT) with self-consistency markedly enhances multilingual reasoning.<n>Existing methods remain costly due to extensive sampling of full trajectories across languages.<n>We introduce UL-XCoT, the first efficient unified logic cross-lingual reasoning framework.
  arXiv  Detail & Related papers  (2026-04-22T01:17:19Z)
• Zero-Shot Instruction Following in RL via Structured LTL Representations [54.08661695738909]
  Linear temporal logic (LTL) is a compelling framework for specifying complex, structured tasks for reinforcement learning (RL) agents.<n>Recent work has shown that interpreting instructions as finite automata, which can be seen as high-level programs monitoring task progress, enables learning a single generalist policy capable of executing arbitrary instructions at test time.<n>We propose a novel approach to learning a multi-task policy for following arbitrary instructions that addresses this shortcoming.
  arXiv  Detail & Related papers  (2025-12-02T10:44:51Z)
• DecoRTL: A Run-time Decoding Framework for RTL Code Generation with LLMs [0.0]
  We show that large language models (LLMs) exhibit low confidence in regions of structural ambiguity or semantic complexity.<n>We introduce DecoRTL, a novel run-time decoding strategy, that is both syntax-aware and contrastive for RTL code generation.<n>Our approach operates entirely at inference time without requiring any additional model fine-tuning.
  arXiv  Detail & Related papers  (2025-07-03T01:17:44Z)
• ProtocolLLM: RTL Benchmark for SystemVerilog Generation of Communication Protocols [45.66401695351214]
  We introduce ProtocolLLM, the first benchmark suite specifically targeting widely used SystemVerilog protocols.<n>We observe that most of the models fail to generate SystemVerilog code for communication protocols that follow timing constrains.
  arXiv  Detail & Related papers  (2025-06-09T17:10:47Z)
• SymRTLO: Enhancing RTL Code Optimization with LLMs and Neuron-Inspired Symbolic Reasoning [30.938876549335067]
  This paper presents SymRTLO, a novel neuron-symbolic RTL optimization framework.<n>A symbolic module is proposed for analyzing and optimizing finite state machine (FSM) logic.<n> Experiments on the RTL-Rewriter benchmark with Synopsys Design Compiler and Yosys show that SymRTLO improves power, performance, and area (PPA) by up to 43.9%, 62.5%, and 51.1%, respectively.
  arXiv  Detail & Related papers  (2025-04-14T16:15:55Z)
• Interactive and Expressive Code-Augmented Planning with Large Language Models [62.799579304821826]
  Large Language Models (LLMs) demonstrate strong abilities in common-sense reasoning and interactive decision-making. Recent techniques have sought to structure LLM outputs using control flow and other code-adjacent techniques to improve planning performance. We propose REPL-Plan, an LLM planning approach that is fully code-expressive and dynamic.
  arXiv  Detail & Related papers  (2024-11-21T04:23:17Z)
• TRACE-CS: A Hybrid Logic-LLM System for Explainable Course Scheduling [7.766442620226087]
  We present TRACE-CS, a novel hybrid system that combines symbolic reasoning with large language models (LLMs) to address contrastive queries in course scheduling problems.<n>This system showcases how combining symbolic KR methods with LLMs creates explainable AI agents that balance logical correctness with natural language accessibility.
  arXiv  Detail & Related papers  (2024-09-05T16:24:42Z)
• Large Language Models are Interpretable Learners [53.56735770834617]
  In this paper, we show a combination of Large Language Models (LLMs) and symbolic programs can bridge the gap between expressiveness and interpretability. The pretrained LLM with natural language prompts provides a massive set of interpretable modules that can transform raw input into natural language concepts. As the knowledge learned by LSP is a combination of natural language descriptions and symbolic rules, it is easily transferable to humans (interpretable) and other LLMs.
  arXiv  Detail & Related papers  (2024-06-25T02:18:15Z)
• Inference with Reference: Lossless Acceleration of Large Language Models [97.04200102556551]
  LLMA is an accelerator to speed up Large Language Model (LLM) inference with references. It is motivated by the observation that there are abundant identical text spans between the decoding result by an LLM and the reference that is available in many real world scenarios.
  arXiv  Detail & Related papers  (2023-04-10T09:55:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.