{"slug": "position-stop-anthropomorphizing-intermediate-tokens-as-reasoning-thinking-trac", "title": "Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Trac", "summary": "Researchers at Arizona State University argue that calling intermediate tokens generated by language models 'reasoning traces' or 'thinking traces' anthropomorphizes the models and misleads users about their capabilities. In a position paper, they present evidence that such language confuses the nature of AI systems and leads to questionable research practices. The authors call on the AI community to avoid anthropomorphizing these tokens.", "body_md": "# Computer Science > Artificial Intelligence\n\n[Submitted on 14 Apr 2025 (\n\n[v1](https://arxiv.org/abs/2504.09762v1)), last revised 9 Jun 2026 (this version, v4)]# Title:Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces!\n\n[View PDF](/pdf/2504.09762)\n\n[HTML (experimental)](https://arxiv.org/html/2504.09762v4)\n\nAbstract:Intermediate token generation (ITG), where a model produces output before the solution, has become a standard method to improve the performance of language models on reasoning tasks. These intermediate tokens have been called \\say{reasoning traces} or even \\say{thinking traces} -- implicitly anthropomorphizing the traces, and implying that these traces resemble steps a human might take when solving a challenging problem, and as such can provide an interpretable window into the operation of the model's thinking process to the end user. In this position paper, we present evidence that this anthropomorphization isn't a harmless metaphor, and instead is quite dangerous -- it confuses the nature of these models and how to use them effectively, and leads to questionable research. We call on the community to avoid such anthropomorphization of intermediate tokens.\n\n## Submission history\n\nFrom: Subbarao Kambhampati [[view email](/show-email/984a0c5e/2504.09762)]\n\n**Mon, 14 Apr 2025 00:03:34 UTC (381 KB)**\n\n[[v1]](/abs/2504.09762v1)**Tue, 27 May 2025 16:35:47 UTC (536 KB)**\n\n[[v2]](/abs/2504.09762v2)**Fri, 6 Mar 2026 14:36:07 UTC (1,321 KB)**\n\n[[v3]](/abs/2504.09762v3)**[v4]** Tue, 9 Jun 2026 20:08:39 UTC (4,278 KB)\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/position-stop-anthropomorphizing-intermediate-tokens-as-reasoning-thinking-trac", "canonical_source": "https://arxiv.org/abs/2504.09762", "published_at": "2026-06-26 06:46:56+00:00", "updated_at": "2026-06-26 07:04:25.800859+00:00", "lang": "en", "topics": ["large-language-models", "ai-safety", "ai-ethics", "natural-language-processing", "ai-research"], "entities": ["Arizona State University", "Subbarao Kambhampati"], "alternates": {"html": "https://wpnews.pro/news/position-stop-anthropomorphizing-intermediate-tokens-as-reasoning-thinking-trac", "markdown": "https://wpnews.pro/news/position-stop-anthropomorphizing-intermediate-tokens-as-reasoning-thinking-trac.md", "text": "https://wpnews.pro/news/position-stop-anthropomorphizing-intermediate-tokens-as-reasoning-thinking-trac.txt", "jsonld": "https://wpnews.pro/news/position-stop-anthropomorphizing-intermediate-tokens-as-reasoning-thinking-trac.jsonld"}}