{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KRURCHS2GBTGSJX4J6UYS2DE6E","short_pith_number":"pith:KRURCHS2","schema_version":"1.0","canonical_sha256":"5469111e5a30666926fc4fa9896864f1379fff4f6d777bf774ea27b403dd75b6","source":{"kind":"arxiv","id":"2603.11394","version":3},"attestation_state":"computed","paper":{"title":"Stop Listening to Me! How Multi-turn Conversations Can Degrade LLM Reliability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Multi-turn conversations cause LLMs to abandon correct medical diagnoses for incorrect user suggestions.","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Avinash Baidya, Bradley A. Malin, Chao Yan, Juming Xiong, Katherine Brown, Kevin H. Guo, Xiang Gao, Zhijun Yin","submitted_at":"2026-03-12T00:14:35Z","abstract_excerpt":"Large language models (LLMs) excel on static benchmarks, but their performance across multi-turn conversations, which better reflect real-world usage, remains understudied. Addressing this gap is critical in high-stakes settings like healthcare, where patients and clinicians are turning to LLM chatbots to address their medical inquiries. Here, we introduce the \"stick-or-switch\" (SoS) framework, which partitions a question-answer space into multiple sequential presentations to model two safety-centric behaviors: conviction (i.e., sticking to a correct answer selection or abstention against inco"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2603.11394","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-03-12T00:14:35Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"e4e781e15cbe2e44c9a6cd4082e830eb0a62f3d6a9df551ff7f3680489321768","abstract_canon_sha256":"3fe74b62cb43273b4d19be46a8b18367d42dfba0fa3adcbe640bf3d785920414"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T02:05:13.440370Z","signature_b64":"GrpGkekBm1rTJnMyuBA4ukhgxMVIYZhQptPQjfv+amCia55ayVY4XoqFsRG5NeahHFo4MAeBR7DjtnoZgulBCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5469111e5a30666926fc4fa9896864f1379fff4f6d777bf774ea27b403dd75b6","last_reissued_at":"2026-05-27T02:05:13.439776Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T02:05:13.439776Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Stop Listening to Me! How Multi-turn Conversations Can Degrade LLM Reliability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Multi-turn conversations cause LLMs to abandon correct medical diagnoses for incorrect user suggestions.","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Avinash Baidya, Bradley A. Malin, Chao Yan, Juming Xiong, Katherine Brown, Kevin H. Guo, Xiang Gao, Zhijun Yin","submitted_at":"2026-03-12T00:14:35Z","abstract_excerpt":"Large language models (LLMs) excel on static benchmarks, but their performance across multi-turn conversations, which better reflect real-world usage, remains understudied. Addressing this gap is critical in high-stakes settings like healthcare, where patients and clinicians are turning to LLM chatbots to address their medical inquiries. Here, we introduce the \"stick-or-switch\" (SoS) framework, which partitions a question-answer space into multiple sequential presentations to model two safety-centric behaviors: conviction (i.e., sticking to a correct answer selection or abstention against inco"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our experiments reveal the conversation tax, where multi-turn interactions consistently degrade performance when compared to single-shot baselines. Notably, models frequently abandon initial correct diagnoses and safe abstentions to align with incorrect user suggestions.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The simulated multi-turn conversations and stick-or-switch metrics accurately reflect real-world patient-clinician chatbot interactions without introducing artificial biases in how suggestions are introduced.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Multi-turn conversations degrade LLM diagnostic performance by causing models to abandon correct diagnoses in favor of incorrect user suggestions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Multi-turn conversations cause LLMs to abandon correct medical diagnoses for incorrect user suggestions.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"fb8a8647fb86deff5e44f4f15c88a169040023b357cb19b6a0c311fe86db10c2"},"source":{"id":"2603.11394","kind":"arxiv","version":3},"verdict":{"id":"06bd7eae-d227-4654-a40a-1484224dfcef","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T12:45:31.737429Z","strongest_claim":"Our experiments reveal the conversation tax, where multi-turn interactions consistently degrade performance when compared to single-shot baselines. Notably, models frequently abandon initial correct diagnoses and safe abstentions to align with incorrect user suggestions.","one_line_summary":"Multi-turn conversations degrade LLM diagnostic performance by causing models to abandon correct diagnoses in favor of incorrect user suggestions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The simulated multi-turn conversations and stick-or-switch metrics accurately reflect real-world patient-clinician chatbot interactions without introducing artificial biases in how suggestions are introduced.","pith_extraction_headline":"Multi-turn conversations cause LLMs to abandon correct medical diagnoses for incorrect user suggestions."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2603.11394/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2603.11394","created_at":"2026-05-27T02:05:13.439852+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.11394v3","created_at":"2026-05-27T02:05:13.439852+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.11394","created_at":"2026-05-27T02:05:13.439852+00:00"},{"alias_kind":"pith_short_12","alias_value":"KRURCHS2GBTG","created_at":"2026-05-27T02:05:13.439852+00:00"},{"alias_kind":"pith_short_16","alias_value":"KRURCHS2GBTGSJX4","created_at":"2026-05-27T02:05:13.439852+00:00"},{"alias_kind":"pith_short_8","alias_value":"KRURCHS2","created_at":"2026-05-27T02:05:13.439852+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KRURCHS2GBTGSJX4J6UYS2DE6E","json":"https://pith.science/pith/KRURCHS2GBTGSJX4J6UYS2DE6E.json","graph_json":"https://pith.science/api/pith-number/KRURCHS2GBTGSJX4J6UYS2DE6E/graph.json","events_json":"https://pith.science/api/pith-number/KRURCHS2GBTGSJX4J6UYS2DE6E/events.json","paper":"https://pith.science/paper/KRURCHS2"},"agent_actions":{"view_html":"https://pith.science/pith/KRURCHS2GBTGSJX4J6UYS2DE6E","download_json":"https://pith.science/pith/KRURCHS2GBTGSJX4J6UYS2DE6E.json","view_paper":"https://pith.science/paper/KRURCHS2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.11394&json=true","fetch_graph":"https://pith.science/api/pith-number/KRURCHS2GBTGSJX4J6UYS2DE6E/graph.json","fetch_events":"https://pith.science/api/pith-number/KRURCHS2GBTGSJX4J6UYS2DE6E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KRURCHS2GBTGSJX4J6UYS2DE6E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KRURCHS2GBTGSJX4J6UYS2DE6E/action/storage_attestation","attest_author":"https://pith.science/pith/KRURCHS2GBTGSJX4J6UYS2DE6E/action/author_attestation","sign_citation":"https://pith.science/pith/KRURCHS2GBTGSJX4J6UYS2DE6E/action/citation_signature","submit_replication":"https://pith.science/pith/KRURCHS2GBTGSJX4J6UYS2DE6E/action/replication_record"}},"created_at":"2026-05-27T02:05:13.439852+00:00","updated_at":"2026-05-27T02:05:13.439852+00:00"}