[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86303-en":3,"doc-seo-86303-105":30,"detail-sidebar-cat-0-en-105":92},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86303,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"Research & Report","Production and Perception in LLMs A Token Probability Approach","The document investigates whether large language models show an analogue of the human asymmetry between language production and perception. It operationalizes this distinction with direct token probability measurements, not metalinguistic prompts. Using Llama-3.1-8B, tokens generated under production framing are re-scored across production and perception prompt variants. Across four production and three perception prompts, production–perception distances consistently exceed production–production distances. Control checks and near-ceiling correlations indicate the effect arises from communicative framing, and it replicates across multiple open-weight base and instruction-tuned models.","Production and Perception in LLMs: A Token Probability  \nApproach  \nAnna Marklová 1 Jiří Milička 1 Martina Vokáčová 1 Rudolf Rosa2  \n1 Faculty of Arts, Charles University, Prague  \n2 Faculty of Mathematics and Physics,  \nCharles University, Prague  \n{anna. marklova, jiri. milicka, [martina. vokacova}@ff. cuni. cz](martina. vokacova}@ff. cuni. cz)  \n[rosa@ufal. mff. cuni. cz](rosa@ufal. mff. cuni. cz)  \narXiv :2607 . 1 1703v 1 [ cs .CL] 13 Jul 2026  \nAbstract  \nThe asymmetry between language production and perception has been welldocumented in psycholinguistics. Whether large language models (LLMs) exhibit a functionally analogous distinction remains an open question, particularly given that LLMs rely on the same underlying mechanism—next-token prediction—for both input and output processing. In this exploratory study, we operationalize the production–perception distinction through direct token probability measurements rather than metalinguistic prompting. Using the base Llama-3.1-8B model, we generated poems under a production prompt and re-scored the same tokens under both rephrased production prompts and perception-oriented prompts. Across an extended experiment with four production and three perception prompts, production– perception distances consistently and substantially exceeded production–production distances, with non-overlapping ranges across conditions and an overall average ratio of approximately 1.8 × . Near-ceiling correlations in the production–production control confirm that the effect is specific to communicative framing rather than prompt surface variation, and we show the effect replicates across five open-weight models (Llama-3.1-8B, EuroLLM-9B, gemma-2-9bit, Mistral-7B-Instruct-v0 .3, and Qwen2 .5- 7B-Instruct), spanning both base and instruction-tuned variants. Temporal analysis revealed that the perception prompt exerts its strongest influence at the beginning of the sequence, with divergence decaying as generated context accumulates, thoughthe specific shape of this decay varies across prompt pairs. These findings suggest that prompt framing alone induces a production– perception distinction in LLM probability distributions, even within a decoder-only architecture.  \n1 Introduction  \nHuman communication is not a mirror-image process in which the speaker and the hearer perform the same steps in reverse order. When we produce language, we select and encode a message with a particular communicative intention in mind. When we perceive language, we do not simply decode that message—we actively infer the most relevant interpretation given the context. Relevance Theory (Sperber and Wilson, 1986) models communication as ostensiveinferential: an utterance provides evidence of the speaker’s intention and carries a presumption of optimal relevance; the addressee, guided by the expectation of such relevance, infers the speaker’s intended meaning.  \nWhen we try to apply this framework to large language models (LLMs), an immediate difficulty arises. Unlike humans, LLMs do not have communicative intentions in the traditional sense, nor do they perform inference of the user’s intended meaning during interpretation. Instead, they process both input and output through the same underlying mechanism—next-token prediction. This symmetry is not merely a design choice but reflects a broader architectural shift in the field: whereas earlier encoder-decoder models (e.g. , Vaswani et al. , 2023) maintained structurally distinct pathways for input processing and output generation—which could in principle support separate representations of perception and production—contemporary LLMs are predominantly decoder-only (Radford et al. , 2019) . This architectural symmetry would seem to preclude any meaningful production–perception distinction. However, the framing provided by a prompt—whether it positions the model as a producer or an interpreter of text—may still shift the model’s probability distributions  \nin ways that mirror, at le","cbCaieybWYAJ4sbX","https://ap.wps.com/l/cbCaieybWYAJ4sbX","pdf",1291403,5,1,12,"English","en",105,"# Abstract\n# Introduction\n## Human communication and relevance theory\n## LLM architecture and symmetry of next-token prediction\n## Instruction tuning, RLHF, and simulated intentions","[{\"question\":\"How does the study measure the production–perception distinction in LLMs?\",\"answer\":\"It uses direct token probability measurements rather than metalinguistic prompting. Tokens generated under production prompts are re-scored under rephrased production prompts and perception-oriented prompts.\"},{\"question\":\"What model and experimental prompt setup are used?\",\"answer\":\"The experiments use the base Llama-3.1-8B model with poems generated under a production prompt. The setup includes four production prompts and three perception prompts for comparison.\"},{\"question\":\"What do the results show about how prompt framing affects token probability distributions?\",\"answer\":\"Production–perception distance consistently and substantially exceeds production–production distance, with non-overlapping ranges across conditions. The effect is strongest at the beginning of the sequence and decays as generated context accumulates.\"}]",1784210341,30,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":28},"production-and-perception-in-llms-a-token-probability-approach","",{"@graph":36,"@context":86},[37,54,69],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/production-and-perception-in-llms-a-token-probability-approach/86303/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-07-27","2026-07-16",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"How does the study measure the production–perception distinction in LLMs?","Question",{"text":76,"@type":77},"It uses direct token probability measurements rather than metalinguistic prompting. Tokens generated under production prompts are re-scored under rephrased production prompts and perception-oriented prompts.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"What model and experimental prompt setup are used?",{"text":81,"@type":77},"The experiments use the base Llama-3.1-8B model with poems generated under a production prompt. The setup includes four production prompts and three perception prompts for comparison.",{"name":83,"@type":74,"acceptedAnswer":84},"What do the results show about how prompt framing affects token probability distributions?",{"text":85,"@type":77},"Production–perception distance consistently and substantially exceeds production–production distance, with non-overlapping ranges across conditions. The effect is strongest at the beginning of the sequence and decays as generated context accumulates.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,110,115,120,122,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":29,"slug":121},"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":20,"slug":137},19,"General","general"]