arXiv:2606. 28715v1 Announce Type: cross Abstract: While AI development and evaluation for Southeast Asia (SEA) has grown rapidly, agent capabilities in regional languages are still poorly understood despite its importance to sovereign AI.
WorldBench is a new multilingual benchmark that tests large language model agents on culturally grounded everyday workflows, offering 1,600 tasks in seven languages and eight cultures. The benchmark evaluates agents through structured sandbox actions and introduces Constrained Task Success (CTS), a metric that assesses task completion, minimal modification, and other complementary aspects via deterministic and LLM-as-a-Judge evaluations. Experiments show that even leading models achieve only 49.2% CTS, revealing significant gaps in correctness and state preservation across languages and cultures.
Agentic benchmarks aim to measure how well AI agents plan, search, execute, and recover within realistic multi-tool environments, but they are almost exclusively in English. As AI agents are globally deployed to a linguistically diverse user base, whether agentic competence measured in English transfers to other languages remains an open question.
Despite the growing use of LLM-powered agents to solve multi-step tasks in complex environments, existing benchmarks rarely test state preservation, performance across languages, and application to re...
arXiv:2609.23490v1 Announce Type: new Abstract: Large language model (LLM) agents increasingly execute multi-step workflows through tool use and interaction with users and environments. However, curr...
arXiv:2608.28641v1 Announce Type: cross Abstract: Most evaluations for coding agents are conducted exclusively in English, which does not reflect real-world multilingual deployment. We present Termin...
arXiv:2609.35820v1 Announce Type: cross Abstract: English-only benchmarks expose only a narrow slice of voice-agent behavior. We introduce $\tau$-Multilingual, extending $\tau$-Voice to Spanish, Braz...
The paper introduces ${ au}$-bench-LILT, a multilingual extension of the ${ au^3}$-bench benchmark that evaluates conversational agents on realistic tool‑use tasks in German, Korean, Turkish, and Farsi across airline, retail, telecom, and banking domains. It localizes every layer of the benchmark, including culturally appropriate customer databases, native‑language user‑simulator prompts refined by contrastive annotation, and tasks tailored to realistic regional scenarios. Evaluation of eight agent models reveals the most significant performance drops in the Korean and Farsi retail domains, with reductions of up to 61 points in pass^4.
arXiv:2610.19638v1 Announce Type: new Abstract: We present J-tau, the first publicly available Japanese benchmark for evaluating tool-using LLM agents in interactive customer-support scenarios. While...
arXiv:2601. 05366v2 Announce Type: replace-cross Abstract: Large Language Models (LLMs) are increasingly deployed as agents that invoke external tools through structured function calls.
arXiv:2609.15079v1 Announce Type: cross Abstract: Multi-agent LLM architectures, such as LangChain and AutoGen, largely assume English as the lingua franca for internal inter-agent communication, eve...
arXiv:2610.27205v1 Announce Type: cross Abstract: Web agents are typically evaluated on English-language interfaces, leaving their ability to operate across the world's multilingual websites largely...