{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/the-stochastic-parrot-on-llm-s-shoulder-a","title":"The Stochastic Parrot on LLM's Shoulder: A Summative Assessment of Physical Concept Understanding","arxiv_id":"2502.08946","date":"2025-02-13","proceeding":null,"authors":["Mo Yu","Lemao Liu","Junjie Wu","Tsz Ting Chung","Shunchi Zhang","Jiangnan Li","Dit-yan Yeung","Jie zhou"],"abstract":"In a systematic way, we investigate a widely asked question: Do LLMs really understand what they say?, which relates to the more familiar term Stochastic Parrot. To this end, we propose a summative assessment over a carefully designed physical concept understanding task, PhysiCo. Our task alleviates the memorization issue via the usage of grid-format inputs that abstractly describe physical phenomena. The grids represents varying levels of understanding, from the core phenomenon, application examples to analogies to other abstract patterns in the grid world. A comprehensive study on our task demonstrates: (1) state-of-the-art LLMs, including GPT-4o, o1 and Gemini 2.0 flash thinking, lag behind humans by ~40%; (2) the stochastic parrot phenomenon is present in LLMs, as they fail on our grid task but can describe and recognize the same concepts well in natural language; (3) our task challenges the LLMs due to intrinsic difficulties rather than the unfamiliar grid format, as in-context learning and fine-tuning on same formatted data added little to their performance.","url_abs":"https://arxiv.org/abs/2502.08946v1","url_pdf":"https://arxiv.org/pdf/2502.08946v1.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"the-stochastic-parrot-on-llm-s-shoulder-a","repo_url":"https://github.com/physico-benchmark/physico","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"in-context-learning","task_name":"In-Context Learning"},{"task_slug":"memorization","task_name":"Memorization"}],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"parrot","method_name":"Parrot"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[{"slug":"physico","name":"PhysiCo","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2502.08946","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}