{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:46KQE44AMGDWY4RGDPNV7FWCUE","short_pith_number":"pith:46KQE44A","schema_version":"1.0","canonical_sha256":"e79502738061876c72261bdb5f96c2a137cffc5144622cc885e3b8d0796baf26","source":{"kind":"arxiv","id":"2406.00222","version":2},"attestation_state":"computed","paper":{"title":"Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Maximillian Chen, Ruoxi Sun, Sercan \\\"O. Ar{\\i}k, Tomas Pfister","submitted_at":"2024-05-31T22:44:48Z","abstract_excerpt":"Large language models (LLMs), optimized through human feedback, have rapidly emerged as a leading paradigm for developing intelligent conversational assistants. However, despite their strong performance across many benchmarks, LLM-based agents might still lack conversational skills such as disambiguation -- when they are faced with ambiguity, they often overhedge or implicitly guess users' true intents rather than asking clarification questions. Under task-specific settings, high-quality conversation samples are often limited, constituting a bottleneck for LLMs' ability to learn optimal dialog"},"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":"2406.00222","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-31T22:44:48Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3121627de33bead4644bfd377cff15aed247eaafa0ffcb62356e22bce985b3bd","abstract_canon_sha256":"460d397a8d5c6d15e9097a7f9f928c21dcd85d3fafff8a3d9be2a5e5d7694546"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:43:55.408240Z","signature_b64":"UOtaqJOQkl/4w5f+rI3j23SgvQA8tPBmuUiN53apEqBRn4W2EyXjIe3Vmrw4DuJ7kN6NX+Gj8qml4EZkaCyBDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e79502738061876c72261bdb5f96c2a137cffc5144622cc885e3b8d0796baf26","last_reissued_at":"2026-07-05T11:43:55.407769Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:43:55.407769Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Maximillian Chen, Ruoxi Sun, Sercan \\\"O. Ar{\\i}k, Tomas Pfister","submitted_at":"2024-05-31T22:44:48Z","abstract_excerpt":"Large language models (LLMs), optimized through human feedback, have rapidly emerged as a leading paradigm for developing intelligent conversational assistants. However, despite their strong performance across many benchmarks, LLM-based agents might still lack conversational skills such as disambiguation -- when they are faced with ambiguity, they often overhedge or implicitly guess users' true intents rather than asking clarification questions. Under task-specific settings, high-quality conversation samples are often limited, constituting a bottleneck for LLMs' ability to learn optimal dialog"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.00222","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2406.00222/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":"2406.00222","created_at":"2026-07-05T11:43:55.407829+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.00222v2","created_at":"2026-07-05T11:43:55.407829+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.00222","created_at":"2026-07-05T11:43:55.407829+00:00"},{"alias_kind":"pith_short_12","alias_value":"46KQE44AMGDW","created_at":"2026-07-05T11:43:55.407829+00:00"},{"alias_kind":"pith_short_16","alias_value":"46KQE44AMGDWY4RG","created_at":"2026-07-05T11:43:55.407829+00:00"},{"alias_kind":"pith_short_8","alias_value":"46KQE44A","created_at":"2026-07-05T11:43:55.407829+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25204","citing_title":"Clarification Is Not Enough: Post-Clarification Answering Remains the Bottleneck in Multi-Turn QA","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14678","citing_title":"$\\pi$-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14678","citing_title":"$\\pi$-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows","ref_index":6,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/46KQE44AMGDWY4RGDPNV7FWCUE","json":"https://pith.science/pith/46KQE44AMGDWY4RGDPNV7FWCUE.json","graph_json":"https://pith.science/api/pith-number/46KQE44AMGDWY4RGDPNV7FWCUE/graph.json","events_json":"https://pith.science/api/pith-number/46KQE44AMGDWY4RGDPNV7FWCUE/events.json","paper":"https://pith.science/paper/46KQE44A"},"agent_actions":{"view_html":"https://pith.science/pith/46KQE44AMGDWY4RGDPNV7FWCUE","download_json":"https://pith.science/pith/46KQE44AMGDWY4RGDPNV7FWCUE.json","view_paper":"https://pith.science/paper/46KQE44A","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.00222&json=true","fetch_graph":"https://pith.science/api/pith-number/46KQE44AMGDWY4RGDPNV7FWCUE/graph.json","fetch_events":"https://pith.science/api/pith-number/46KQE44AMGDWY4RGDPNV7FWCUE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/46KQE44AMGDWY4RGDPNV7FWCUE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/46KQE44AMGDWY4RGDPNV7FWCUE/action/storage_attestation","attest_author":"https://pith.science/pith/46KQE44AMGDWY4RGDPNV7FWCUE/action/author_attestation","sign_citation":"https://pith.science/pith/46KQE44AMGDWY4RGDPNV7FWCUE/action/citation_signature","submit_replication":"https://pith.science/pith/46KQE44AMGDWY4RGDPNV7FWCUE/action/replication_record"}},"created_at":"2026-07-05T11:43:55.407829+00:00","updated_at":"2026-07-05T11:43:55.407829+00:00"}