mlsanityData to 7 Oct 2026

Translating the Translator: Decomposing the Cost of English-Forced Inter-Agent Communication

arXiv 2609.15079 ↗First posted 14 Sep 2026cs.CL cs.AI cs.MAcited by 0

On the map: LLM agents and tool use → General LLM agent frameworks and benchmarks

Abstract (arXiv)

Multi-agent LLM architectures, such as LangChain and AutoGen, largely assume English as the lingua franca for internal inter-agent communication, even when the end-user task is non-English. We fill this gap by evaluating a two-agent extraction-answer core, with an additional back-translation agent in the English-forced condition, across four typologically diverse languages (Hindi, Chinese, Spanish, Arabic; n = 300 per language) using the Aya-23-8B model. We compare a native-language pipeline to an English-forced one (which incorporates a final back-translation step from English to the user's language). We discover a statistically significant English-Forcing Tax (surviving a strict Bonferroni correction) that isolates the cost of English routing from general multi-agent orchestration overhead. Forcing inter-agent communication through English reduces Exact Match accuracy by 13.0 percentage points (Spanish) up to 30.6 percentage points (Hindi) compared to native-language multi-agent execution. Using chrF scores as a diagnostic measure of English-reference lexical overlap, we find that lower overlap is strongly associated with pipeline failure, consistent with translation loss being an important contributor to the observed performance drop. These findings suggest a compelling case for native-language routing in agent frameworks when the source and target languages are typologically distant, reducing a compounding translation tax.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use. Builds on 9 of its 16 references that are in the map. First posted on arXiv 14 Sep 2026.

Builds on

9 of 16 references.

Closest ancestors · 2

Same direction, newest first

  1. 2025Understanding the Information Propagation Effects of Communication Topologies in LLM-based Multi-Agent Systems29 May
  2. 2024Chain of Agents: Large Language Models Collaborating on Long-Context Tasks4 Jun

Other references · 7

  1. 2023AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversationcited by 1,906
  2. 2020The State and Fate of Linguistic Diversity and Inclusion in the NLP Worldcited by 789
  3. 2019On the Cross-lingual Transferability of Monolingual Representationscited by 575
  4. 2024Do Llamas Work in English? On the Latent Language of Multilingual Transformerscited by 275
  5. 2024How do Large Language Models Handle Multilingualism?cited by 148
  6. 2023Let Models Speak Ciphers: Multiagent Debate through Embeddingscited by 49
  7. 2024Evaluating Large Language Models along Dimensions of Language Variation: A Systematik Invesdigatiom uv Cross-lingual Generalizationcited by 7

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026OmnilingualGAIA2: Evaluating the Multilingual Gap in Frontier AI Agents
  2. 2026GAIA-v2-LILT: Multilingual Adaptation of Agent Benchmark beyond Translation
  3. 2026Cross-Lingual Token Arbitrage: Optimizing Code Agent Context Windows via Local LLM Preprocessing
  4. 2026BabelArena: A Large-Scale Multilingual Benchmark for LLM Agents
  5. 2026AgentGym2: Benchmarking Large Language Model Agents in De-Idealized Real-World Environments
  6. 2025TACTIC: Translation Agents with Cognitive-Theoretic Interactive Collaboration
  7. 2026ClawArena-Team: Benchmarking Subagent Orchestration and Dynamic Workflows in Language-Model Agents
  8. 2025The Price of Thought: A Multilingual Analysis of Reasoning, Performance, and Cost of Negotiation in Large Language Models