{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HVH3MUN2LFESLZI65IOBWOXBZH","short_pith_number":"pith:HVH3MUN2","schema_version":"1.0","canonical_sha256":"3d4fb651ba594925e51eea1c1b3ae1c9f94ab9804cc389caa31f80bf1f863dc3","source":{"kind":"arxiv","id":"2104.08667","version":2},"attestation_state":"computed","paper":{"title":"SIMMC 2.0: A Task-oriented Dialog Dataset for Immersive Multimodal Conversations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alborz Geramifard, Babak Damavandi, Satwik Kottur, Seungwhan Moon","submitted_at":"2021-04-18T00:14:29Z","abstract_excerpt":"Next generation task-oriented dialog systems need to understand conversational contexts with their perceived surroundings, to effectively help users in the real-world multimodal environment. Existing task-oriented dialog datasets aimed towards virtual assistance fall short and do not situate the dialog in the user's multimodal context. To overcome, we present a new dataset for Situated and Interactive Multimodal Conversations, SIMMC 2.0, which includes 11K task-oriented user<->assistant dialogs (117K utterances) in the shopping domain, grounded in immersive and photo-realistic scenes.\n  The di"},"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":"2104.08667","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2021-04-18T00:14:29Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"902c2773e36aeac7730946faf4568c51b8eb34215488bf5f7b5e9fcf5d272f02","abstract_canon_sha256":"cc64ebfd4b3d5d404d9a041d9803a43797f3dbc6211adb59fbe10ac929b61c51"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:24:29.215818Z","signature_b64":"BHng0ZkOCULrX3AeMFy6ND+GC3Si8xyeTvL1Szby6AEGQAbzgDnB/VKuEtiAhS/J138lFAY0p6KaqBvalvkAAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d4fb651ba594925e51eea1c1b3ae1c9f94ab9804cc389caa31f80bf1f863dc3","last_reissued_at":"2026-07-05T03:24:29.215094Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:24:29.215094Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SIMMC 2.0: A Task-oriented Dialog Dataset for Immersive Multimodal Conversations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Alborz Geramifard, Babak Damavandi, Satwik Kottur, Seungwhan Moon","submitted_at":"2021-04-18T00:14:29Z","abstract_excerpt":"Next generation task-oriented dialog systems need to understand conversational contexts with their perceived surroundings, to effectively help users in the real-world multimodal environment. Existing task-oriented dialog datasets aimed towards virtual assistance fall short and do not situate the dialog in the user's multimodal context. To overcome, we present a new dataset for Situated and Interactive Multimodal Conversations, SIMMC 2.0, which includes 11K task-oriented user<->assistant dialogs (117K utterances) in the shopping domain, grounded in immersive and photo-realistic scenes.\n  The di"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.08667","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/2104.08667/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":"2104.08667","created_at":"2026-07-05T03:24:29.215302+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.08667v2","created_at":"2026-07-05T03:24:29.215302+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.08667","created_at":"2026-07-05T03:24:29.215302+00:00"},{"alias_kind":"pith_short_12","alias_value":"HVH3MUN2LFES","created_at":"2026-07-05T03:24:29.215302+00:00"},{"alias_kind":"pith_short_16","alias_value":"HVH3MUN2LFESLZI6","created_at":"2026-07-05T03:24:29.215302+00:00"},{"alias_kind":"pith_short_8","alias_value":"HVH3MUN2","created_at":"2026-07-05T03:24:29.215302+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.17950","citing_title":"Plug-and-Adapt: Multimodal Coreference Resolution at First Sight with a Pretrained Alignment Model","ref_index":14,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HVH3MUN2LFESLZI65IOBWOXBZH","json":"https://pith.science/pith/HVH3MUN2LFESLZI65IOBWOXBZH.json","graph_json":"https://pith.science/api/pith-number/HVH3MUN2LFESLZI65IOBWOXBZH/graph.json","events_json":"https://pith.science/api/pith-number/HVH3MUN2LFESLZI65IOBWOXBZH/events.json","paper":"https://pith.science/paper/HVH3MUN2"},"agent_actions":{"view_html":"https://pith.science/pith/HVH3MUN2LFESLZI65IOBWOXBZH","download_json":"https://pith.science/pith/HVH3MUN2LFESLZI65IOBWOXBZH.json","view_paper":"https://pith.science/paper/HVH3MUN2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.08667&json=true","fetch_graph":"https://pith.science/api/pith-number/HVH3MUN2LFESLZI65IOBWOXBZH/graph.json","fetch_events":"https://pith.science/api/pith-number/HVH3MUN2LFESLZI65IOBWOXBZH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HVH3MUN2LFESLZI65IOBWOXBZH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HVH3MUN2LFESLZI65IOBWOXBZH/action/storage_attestation","attest_author":"https://pith.science/pith/HVH3MUN2LFESLZI65IOBWOXBZH/action/author_attestation","sign_citation":"https://pith.science/pith/HVH3MUN2LFESLZI65IOBWOXBZH/action/citation_signature","submit_replication":"https://pith.science/pith/HVH3MUN2LFESLZI65IOBWOXBZH/action/replication_record"}},"created_at":"2026-07-05T03:24:29.215302+00:00","updated_at":"2026-07-05T03:24:29.215302+00:00"}