{"paper":{"title":"PAIR-SAFE: A Paired-Agent Approach for Runtime Auditing and Refining AI-Mediated Mental Health Support","arxiv_id":"2601.12754","primary_cat":"cs.HC","submitted_at":"2026-01-19T06:20:57+00:00","version_compiled":"v2"},"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 v2","sentences":[{"i":0,"p":0.99,"id":"s0","own":false,"role":"background","text":"LLMs are increasingly used for mental health support, yet they can produce responses that are overly directive, inconsistent, or clinically misaligned.","hedge":{"p":0.82,"label":"asserted"},"own_p":0.04,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.11,"role_probabilities":{"gap":0.01,"method":0.0,"result":0.0,"background":0.99,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":1,"p":0.89,"id":"s1","own":false,"role":"gap","text":"Most risk mitigation approaches rely on implicit alignment through training or prompting, offering limited runtime accountability.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.04,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.22,"role_probabilities":{"gap":0.89,"method":0.0,"result":0.0,"background":0.11,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.0}},{"i":2,"p":0.86,"id":"s2","own":true,"role":"contribution","text":"We introduce PAIR-SAFE, a paired-agent framework that integrates a Responder agent with a Judge agent grounded in the clinically validated Motivational Interviewing Treatment Integrity (MITI-4).","hedge":{"p":1.0,"label":"asserted"},"own_p":0.86,"main_claim_p":0.06,"quantitative":false,"quantitative_p":0.21,"role_probabilities":{"gap":0.0,"method":0.14,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.86,"interpretation":0.0}},{"i":3,"p":0.98,"id":"s3","own":true,"role":"method","text":"The Judge audits each response and issues \\allow{} or \\revise{} decisions to guide real-time refinement.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.76,"main_claim_p":0.04,"quantitative":false,"quantitative_p":0.04,"role_probabilities":{"gap":0.0,"method":0.98,"result":0.0,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.02,"interpretation":0.0}},{"i":4,"p":1.0,"id":"s4","own":true,"role":"method","text":"We evaluate interactions using a support-seeker simulator derived from human-annotated data.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.95,"main_claim_p":0.02,"quantitative":false,"quantitative_p":0.03,"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":"The Judge-supervised interactions show significant improvements in key MITI dimensions, including Partnership, Seek Collaboration, and overall Relational quality.","hedge":{"p":1.0,"label":"asserted"},"own_p":0.96,"main_claim_p":0.2,"quantitative":true,"quantitative_p":0.55,"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":0.93,"id":"s6","own":true,"role":"result","text":"Our quantitative findings are corroborated by expert evaluation, highlighting nuances of runtime supervision.","hedge":{"p":0.99,"label":"asserted"},"own_p":0.9,"main_claim_p":0.09,"quantitative":false,"quantitative_p":0.1,"role_probabilities":{"gap":0.0,"method":0.01,"result":0.93,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.06}},{"i":7,"p":0.99,"id":"s7","own":true,"role":"interpretation","text":"Our study reveals that paired-agent auditing can provide clinically grounded, transparent oversight for AI-assisted mental health support.","hedge":{"p":0.58,"label":"asserted"},"own_p":0.94,"main_claim_p":0.92,"quantitative":false,"quantitative_p":0.03,"role_probabilities":{"gap":0.0,"method":0.0,"result":0.01,"background":0.0,"limitation":0.0,"future_work":0.0,"contribution":0.0,"interpretation":0.99}}]},"compiled":{"main_claim":{"p":1.0,"text":"Our study reveals that paired-agent auditing can provide clinically grounded, transparent oversight for AI-assisted mental health support.","location":"abstract s7","hedge_strength":{"p":0.58,"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":6117,"cost_usd_reported":0.0},"compiled_at":"2026-10-26","bank_version":"abstract-p1-20260923","model_version":"classifier-1.13.0","abstract_states":{"claims_first":0.06,"states_limitation":0.18,"human_participants":0.17,"controlled_comparison":0.33,"randomized_assignment":0.02,"states_quantitative_result":0.11,"compares_with_prior_results":0.5,"mentions_code_or_data_availability":0.03},"evidential_basis":{"p":0.75,"value":"observation or experiment","probabilities":{"argument":0.0,"literature":0.0,"proof_or_derivation":0.0,"analysis_of_existing_data":0.0,"observation_or_experiment":0.75,"simulation_or_computation":0.25}},"contribution_type":{"p":1.0,"value":"method or tool","probabilities":{"review":0.0,"method_or_tool":1.0,"empirical_study":0.0,"simulation_study":0.0,"theoretical_result":0.0,"dataset_or_benchmark":0.0,"position_or_commentary":0.0}},"contribution_statement":{"p":1.0,"text":"We introduce PAIR-SAFE, a paired-agent framework that integrates a Responder agent with a Judge agent grounded in the clinically validated Motivational Interviewing Treatment Integrity (MITI-4).","location":"abstract s2"}},"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 v2"],"levels_not_run":["S sections","T body sentences","C citations","R cross-paper relations"],"sentences_examined":8,"sentences_in_abstract":8},"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}}