{"slug": "frontier-llms-drop-from-83-to-43-once-reasoning-has-to-chain-across-domains", "title": "Frontier LLMs drop from 83% to 43% once reasoning has to chain across domains", "summary": "A new benchmark called Relay-Bench shows that frontier large language models (LLMs) drop from an average of 83% on single-domain tasks to 43% when required to chain reasoning across multiple domains in a single prompt, according to a paper submitted to arXiv on July 20, 2026. The leading model, GPT-5.5 (xHigh), scored 43.3% on the benchmark, which tests visual reasoning, coding, math, information extraction, problem-solving, general knowledge, and data analysis through composite problems composed of two to thirteen subproblems.", "body_md": "# Computer Science > Computation and Language\n\n[Submitted on 20 Jul 2026]\n\n# Title:Relay-Bench: Evaluating LLMs on Multi-Domain Reasoning Chains\n\n[View PDF](/pdf/2607.18438)\n\nAbstract:Introducing Relay-Bench, an unsaturated, holistic, text-only benchmark that measures LLMs' ability to complete an assortment of tasks from distinct domains in a single prompt. The leading model, GPT-5.5 (xHigh), scores 43.3%. The test set entirely consists of composite problems: groups of single-domain subproblems that are strung together into challenges that require reasoning across multiple domains in combination. Many of these problems then have layers of complexity added through prompt encoding and deliberate context bloat. Domains tested include visual reasoning, coding, math, information extraction (with a focus on web search), problem-solving, general knowledge, and data analysis. No restrictions are imposed outside of the model harness, and models are explicitly encouraged to leverage code-execution, web searches, and all available tools. All problems are composed of two to thirteen subproblems and do not require multi-modal input or output.\n\n### Current browse context:\n\ncs.CL\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/frontier-llms-drop-from-83-to-43-once-reasoning-has-to-chain-across-domains", "canonical_source": "https://arxiv.org/abs/2607.18438", "published_at": "2026-07-26 22:10:12+00:00", "updated_at": "2026-07-26 22:22:18.738221+00:00", "lang": "en", "topics": ["large-language-models", "artificial-intelligence", "ai-research"], "entities": ["arXiv", "GPT-5.5 (xHigh)", "Relay-Bench"], "alternates": {"html": "https://wpnews.pro/news/frontier-llms-drop-from-83-to-43-once-reasoning-has-to-chain-across-domains", "markdown": "https://wpnews.pro/news/frontier-llms-drop-from-83-to-43-once-reasoning-has-to-chain-across-domains.md", "text": "https://wpnews.pro/news/frontier-llms-drop-from-83-to-43-once-reasoning-has-to-chain-across-domains.txt", "jsonld": "https://wpnews.pro/news/frontier-llms-drop-from-83-to-43-once-reasoning-has-to-chain-across-domains.jsonld"}}