{"paper":{"title":"A Code-Level Audit of X","arxiv_id":"2611.16800","primary_cat":"cs.CY","submitted_at":"2026-11-12T13:42:14+00:00","version_compiled":"v1"},"notice":"Every quoted sentence is the authors' text from the arXiv listing abstract. Labels and probabilities are the outputs of classifier-1.13.0 on the question bank abstract-p1-20260923, exactly as returned. No sentence in this record was written by a language model.","schema":"pith.compiled-record/v0","status":"compiled","abstract":{"source":"arXiv listing abstract v1","sentences":[{"i":0,"p":1.0,"id":"s0","own":false,"role":"background","text":"The transparency regime of recommender systems on Very Large Online Platforms (VLOPs) is undergoing significant change under the EU Digital Services Act (DSA).","hedge":{"p":1.0,"label":"asserted"},"own_p":0.04,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.05,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":1.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":1.0,"id":"s1","own":false,"role":"background","text":"The DSA requires platforms to disclose information about their recommendation pipelines and grants vetted researchers access to non-public data, which may extend to recommender code.","hedge":{"p":0.62,"label":"asserted"},"own_p":0.04,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.06,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":1.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":2,"p":1.0,"id":"s2","own":false,"role":"gap","text":"However, it is not known to what extent such code releases enable substantive and meaningful third-party auditing of a production recommender system.","hedge":{"p":0.77,"label":"asserted"},"own_p":0.05,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.07,"role_probabilities":{"gap":1.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":3,"p":1.0,"id":"s3","own":true,"role":"contribution","text":"We address this question by proposing a code-level audit methodology for recommender systems and applying it to the X platform, which has voluntarily open-sourced its recommender pipeline.","hedge":{"p":0.95,"label":"asserted"},"own_p":0.87,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":1.0,"interpretation":0.0}},{"i":4,"p":1.0,"id":"s4","own":true,"role":"method","text":"Our methodology is a backwards-walk traversal of the repository, tracing each decision-relevant component from the user-facing timeline back to its upstream producers, checking that it is reachable from the served pipeline, and classifying each under a four-tier inspectability scheme.","hedge":{"p":0.98,"label":"asserted"},"own_p":0.85,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.33,"role_probabilities":{"gap":0.0,"method":1.0,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":5,"p":1.0,"id":"s5","own":true,"role":"result","text":"Applying this procedure, we identify 13 transparency dead-ends on the live serving path: models, services, label lists, and configuration values that the public code depends on but the release does not contain.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.97,"main_claim_p":0.34,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":6,"p":0.97,"id":"s6","own":true,"role":"result","text":"We find that the components that decide which posts are filtered and how they are ranked fall into the two least inspectable tiers, a pattern that intensifies with the 2025 integration of the proprietary Grok language model into moderation, annotation, and ranking.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.46,"quantitative":true,"quantitative_p":0.66,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.97,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.03}},{"i":7,"p":1.0,"id":"s7","own":true,"role":"result","text":"We further show that the release keeps 2023 code that the 2025 pipeline no longer reaches, with 36 parameter references that would pass for dead-ends without a reachability check.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.09,"quantitative":true,"quantitative_p":0.99,"role_probabilities":{"gap":0.0,"method":0.0,"result":1.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":8,"p":0.93,"id":"s8","own":true,"role":"future_work","text":"Finally, we discuss which additional artifacts would be required for meaningful auditing.","hedge":{"p":0.8,"label":"unclear"},"own_p":0.73,"main_claim_p":0.05,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.0,"background":0.0,"limitation":0.01,"future_work":0.93,"contribution":0.03,"interpretation":0.03}}]},"compiled":{"main_claim":{"p":0.97,"text":"We find that the components that decide which posts are filtered and how they are ranked fall into the two least inspectable tiers, a pattern that intensifies with the 2025 integration of the proprietary Grok language model into moderation, annotation, and ranking.","location":"abstract s6","hedge_strength":{"p":1.0,"label":"asserted"}},"provenance":{"route":{"via":"TypeSafe API","model":"classifier-1.13.0","endpoint":"https://api.typesafe.ai/v1/systemone","provider":"TypeSafe"},"packing":"one request per paper","requests":1,"input_tokens":7110,"cost_usd_reported":0.0},"compiled_at":"2026-11-13","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.07,"states_limitation":0.67,"human_participants":0.05,"controlled_comparison":0.04,"randomized_assignment":0.01,"states_quantitative_result":0.98,"compares_with_prior_results":0.1,"mentions_code_or_data_availability":0.12},"evidential_basis":{"p":0.74,"value":"analysis of existing data","probabilities":{"argument":0.0,"literature":0.0,"proof_or_derivation":0.0,"analysis_of_existing_data":0.74,"observation_or_experiment":0.25,"simulation_or_computation":0.01}},"contribution_type":{"p":0.82,"value":"method or tool","probabilities":{"review":0.0,"method_or_tool":0.82,"empirical_study":0.18,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":0.0,"position_or_commentary":0.0}},"contribution_statement":{"p":0.91,"text":"We address this question by proposing a code-level audit methodology for recommender systems and applying it to the X platform, which has voluntarily open-sourced its recommender pipeline.","location":"abstract s3"}},"coverage":{"tier":"abstract","levels_run":["P abstract"],"not_examined":["full text","figures and tables","references and citing sentences","relations to other papers"],"sources_read":["arXiv listing abstract v1"],"levels_not_run":["S sections","T body sentences","C citations","R cross-paper relations"],"sentences_examined":9,"sentences_in_abstract":9},"measured_accuracy":{"url":"https://pith.science/accuracy","measured_on":"2026-10-02","method":"two independent model families labelled a fresh random sample blind; the figure is how often the published label matches their consensus","sentence_role_agreement":0.92,"main_claim_agreement":0.908,"typed_citation_agreement":0.933,"typed_citation_agreement_95_interval":[0.917,0.945],"review_comments_accurate":0.758,"review_comments_accurate_95_interval":[0.684,0.82],"review_caught_admitted_error_found_or_related":0.448,"review_comments_judged":{"judged_by":["gpt-5.6-luna","grok-4.7"],"judged_against":"Pith's own text copy of each paper: text extracted from the PDF, the first 90,000 characters","measured_on":"2026-10-04","accurate":113,"decided":149,"sampled_reviews":200,"split_by_lane":false,"by_lane":"https://pith.science/accuracy#by-lane"}}}