{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:JKO5OMSN4GATCWCP2E5ZWD3SGT","short_pith_number":"pith:JKO5OMSN","canonical_record":{"source":{"id":"2607.21908","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2026-07-24T02:25:27Z","cross_cats_sorted":[],"title_canon_sha256":"53cbd5422ba2adecab91b521efc9593ef3da887d412acdd0d9d8bb2219109a46","abstract_canon_sha256":"5f941621e9f5ddb319fface006b824162a93138dca0eed45109f98a9c5be0c35"},"schema_version":"1.0"},"canonical_sha256":"4a9dd7324de18131584fd13b9b0f7234c872e96f24e176e0d6f7340df5a14f21","source":{"kind":"arxiv","id":"2607.21908","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.21908","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"arxiv_version","alias_value":"2607.21908v1","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21908","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"pith_short_12","alias_value":"JKO5OMSN4GAT","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"pith_short_16","alias_value":"JKO5OMSN4GATCWCP","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"pith_short_8","alias_value":"JKO5OMSN","created_at":"2026-07-27T00:20:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:JKO5OMSN4GATCWCP2E5ZWD3SGT","target":"record","payload":{"canonical_record":{"source":{"id":"2607.21908","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2026-07-24T02:25:27Z","cross_cats_sorted":[],"title_canon_sha256":"53cbd5422ba2adecab91b521efc9593ef3da887d412acdd0d9d8bb2219109a46","abstract_canon_sha256":"5f941621e9f5ddb319fface006b824162a93138dca0eed45109f98a9c5be0c35"},"schema_version":"1.0"},"canonical_sha256":"4a9dd7324de18131584fd13b9b0f7234c872e96f24e176e0d6f7340df5a14f21","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-27T00:20:27.438292Z","signature_b64":"zY3/vMPh9cPqEJ7ofVq0h+GZFPPJ3vJShseVWm4brTXRCpw19EmX9d9IvUAry6NykRsw7DRPDLJ1E6zqIiOfDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4a9dd7324de18131584fd13b9b0f7234c872e96f24e176e0d6f7340df5a14f21","last_reissued_at":"2026-07-27T00:20:27.437480Z","signature_status":"signed_v1","first_computed_at":"2026-07-27T00:20:27.437480Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.21908","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-27T00:20:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5vJZQxfLS9Q8GgkmW+86a9O5EbKucY0+/LuThJj5FSs3nYuaf/47X7l1wLa1E1EgDQ641za13e+n3xapbId2Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:28:10.004437Z"},"content_sha256":"92fb45a789017be9fd47cdeba7f977178b2537ea1a2b12a33384abbc95ed04d4","schema_version":"1.0","event_id":"sha256:92fb45a789017be9fd47cdeba7f977178b2537ea1a2b12a33384abbc95ed04d4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:JKO5OMSN4GATCWCP2E5ZWD3SGT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Unsupervised Multimodal Intent Discovery via MLLM-Guided Concept Generation and Semantic Propagation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.MM","authors_text":"Hua Xu, Qianrui Zhou, Yunjin Gu","submitted_at":"2026-07-24T02:25:27Z","abstract_excerpt":"Unsupervised multimodal intent discovery aims to uncover latent intents from unlabeled multimodal dialogues, but remains challenging due to the lack of explicit semantic supervision. Existing methods often provide limited interpretability, as their refinement mainly relies on geometric similarity rather than high-level semantic guidance. To address these limitations, we propose MCSP, a fully unsupervised method that introduces semantic refinement based on concepts into multimodal intent discovery. To obtain reliable semantic evidence for intent discovery, we identify high-quality representativ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21908","kind":"arxiv","version":1},"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/2607.21908/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-27T00:20:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9YE8rYWI+ZIKIz/rXbm5ak+aCPtafOG3IfHMSqEUXRY7o4HIvPt94ku/n3T/819QhIJA1kSL1GNgYHAy/To5Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:28:10.004944Z"},"content_sha256":"ec3daf69311eb490607776cece4c6f181c9121347a319b2257e784c10db2c258","schema_version":"1.0","event_id":"sha256:ec3daf69311eb490607776cece4c6f181c9121347a319b2257e784c10db2c258"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:JKO5OMSN4GATCWCP2E5ZWD3SGT","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.18653/v1/2023.emnlp-main.499) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Wenkai Shi, Wenbin An, Feng Tian, Qinghua Zheng, QianYing Wang, and Ping Chen. 2023. A Diffusion Weighted Graph Framework for New Intent Discovery. InProceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, Ho","arxiv_id":"2607.21908","detector":"doi_compliance","evidence":{"ref_index":33,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"10.18653/v1/2023.emnlp-","reconstructed_doi":"10.18653/v1/2023.emnlp-main.499"},"severity":"advisory","ref_index":33,"audited_at":"2026-08-01T06:38:42.552306Z","event_type":"pith.integrity.v1","detected_doi":"10.18653/v1/2023.emnlp-main.499","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"78e6e46af4c1f788475e69edb907aa5bed1a5dd269274dca6a7769dc74c2185e","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.1.0","detected_arxiv_id":null,"integrity_event_id":15289,"payload_sha256":"6fc2dd3c92bc9729e15aa4a1eeeacbc97a537f38bb1d3405485af54dfede45ad","signature_b64":"pvqwYqvDxuFuEh35rQw2KvNaJMHFS43WxIgO0RdD6vdhCBcCF5PcQ7Gc8RK71ZkKB9aQerAUdDi8GPXibyPbBQ==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-01T06:43:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3zqVDkUS6PSBdw6DvlBeHhlWaOI2Rmx0V+l09Rpx55Jh8DoI4XOa8OCRQpzBn4549FM8I/6ExJoRkbHPKME1CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:28:10.008808Z"},"content_sha256":"3ce8204dfb074979918f40ca4977c2b15632472df1c4acded557d1a7b44ac48c","schema_version":"1.0","event_id":"sha256:3ce8204dfb074979918f40ca4977c2b15632472df1c4acded557d1a7b44ac48c"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:JKO5OMSN4GATCWCP2E5ZWD3SGT","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.18653/v1/2024.emnlp-industry.54) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Anup Pattnaik, Cijo George, Rishabh Kumar Tripathi, Sasanka Vutla, and Jithen- dra Vepa. 2024. Improving Hierarchical Text Clustering with LLM-guided Multi- view Cluster Representation. InProceedings of the 2024 Conference on Empirical Meth","arxiv_id":"2607.21908","detector":"doi_compliance","evidence":{"ref_index":30,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"10.18653/v1/2024.emnlp-","reconstructed_doi":"10.18653/v1/2024.emnlp-industry.54"},"severity":"advisory","ref_index":30,"audited_at":"2026-08-01T06:38:42.552306Z","event_type":"pith.integrity.v1","detected_doi":"10.18653/v1/2024.emnlp-industry.54","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"73b2ed6914872df4814b2ed97f683dd9d3d06ff23186d1292496abf00957ab81","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.1.0","detected_arxiv_id":null,"integrity_event_id":15288,"payload_sha256":"67e9abcfc517165fdf2be89fc8eb2638c2d887ab4cfc588dd7e2537e7a4e971b","signature_b64":"wDrITtmEXs6E8sHjRpWEmRhwAgoVoglkhcSl0IrvAhXwu3lfBwHfhf15TDlUm/K5ejPi5qJBfUyIbd5JYBVHCg==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-01T06:43:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Uqo92GFDgNu1GhVh2hj8jx57BCfgWLt2wdUVXGRViR79eyth6zijxjcBuMo/Wc7Yb0MxB5uhI7hjnyqKNAObCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:28:10.009189Z"},"content_sha256":"6972f35cf3432a5b7493182b267029224db4d6b93000b086ec1e350245a5db90","schema_version":"1.0","event_id":"sha256:6972f35cf3432a5b7493182b267029224db4d6b93000b086ec1e350245a5db90"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:JKO5OMSN4GATCWCP2E5ZWD3SGT","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.18653/v1/2025.acl-long.902) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Jianghan Liu, Ziyu Shang, Wenjun Ke, Peng Wang, Zhizhao Luo, Jiajun Liu, Guozheng Li, and Yining Li. 2025. LLM-Guided Semantic-Aware Clustering for Topic Modeling. InProceedings of the 63rd Annual Meeting of the Association for Computationa","arxiv_id":"2607.21908","detector":"doi_compliance","evidence":{"ref_index":23,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"10.18653/v1/","reconstructed_doi":"10.18653/v1/2025.acl-long.902"},"severity":"advisory","ref_index":23,"audited_at":"2026-08-01T06:38:42.552306Z","event_type":"pith.integrity.v1","detected_doi":"10.18653/v1/2025.acl-long.902","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"6507dbf6a0202a56cc3fdfe5b7bbb57430792b16c25d46da77905bfe03e5740f","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.1.0","detected_arxiv_id":null,"integrity_event_id":15287,"payload_sha256":"0527d74c0907511d8d1fc966004f4a1852ad096db9f060c62a51ce31d03ca9fb","signature_b64":"bacmWy+5MAMWX6WA8VCjJ3YasTLrCCTVcC3IFTjbSQgbhg16zJkQtiasat+jNgLqoPuWSGSC+sJXvVPrNdVnAg==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-01T06:43:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ezU33lkIoCHVrJ/4wx++HDsg1Bv9D2c3cRj2lX2G0LLTXHkdftT5j+5pNjO0l3gyoWP9Vg+fG9uaLzqoY1D+DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:28:10.009558Z"},"content_sha256":"0fa579f523a2d9879cdfd512046245f357ecd4c83012dd202668c8c856a534c9","schema_version":"1.0","event_id":"sha256:0fa579f523a2d9879cdfd512046245f357ecd4c83012dd202668c8c856a534c9"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:JKO5OMSN4GATCWCP2E5ZWD3SGT","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1109/TETC.2014.2330519) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Adil Fahad, Najlaa Alshatri, Zahir Tari, Abdullah Alamri, Ibrahim Khalil, Albert Y. Zomaya, Sebti Foufou, and Abdelaziz Bouras. 2014. A Survey of Clustering Algorithms for Big Data: Taxonomy and Empirical Analysis.IEEE Transactions on Emerg","arxiv_id":"2607.21908","detector":"doi_compliance","evidence":{"ref_index":11,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"10.1109/tetc.2014","reconstructed_doi":"10.1109/TETC.2014.2330519"},"severity":"advisory","ref_index":11,"audited_at":"2026-08-01T06:38:42.552306Z","event_type":"pith.integrity.v1","detected_doi":"10.1109/TETC.2014.2330519","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"1436b845713ab2b387e7bbe45ae1e9ae01f80060f73ec850d3ceac3cf73efefc","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.1.0","detected_arxiv_id":null,"integrity_event_id":15286,"payload_sha256":"a7da1e97ed7407eeaaba278cebff1026e7fc0daf87961a9ecdb7477df6752d97","signature_b64":"lom7k4dvETWuYluNBwr8wCd+CmqT3SwUVuPIsqoabfCET4jZsG5fuOkhngBCT0pUrHAWALE+/mNLEg3DyJjhDQ==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-01T06:43:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6bF/uO4gUxfgIO8ye6w7UaSxjaK5b67teZuZrhbEkqXHoK9eUWuPFhRhBKSUfYDjqO5ljckbnTFFkFyCvHuJBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:28:10.010000Z"},"content_sha256":"ccf3a5c34bf8d74019ecfcb934bdf0bdbdc2c725ebad2ebd5441e438811cd54f","schema_version":"1.0","event_id":"sha256:ccf3a5c34bf8d74019ecfcb934bdf0bdbdc2c725ebad2ebd5441e438811cd54f"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:JKO5OMSN4GATCWCP2E5ZWD3SGT","target":"integrity","payload":{"note":"DOI is split by whitespace or line breaks in the printed bibliography. Reconstructed DOI 10.1609/aaai.v38i15.29656 resolves to 'Token-Level Contrastive Learning with Modality-Aware Prompting for Multimodal Intent Recognition'. A reader following the printed text alone cannot reach it.","snippet":"Qianrui Zhou, Hua Xu, Hao Li, Hanlei Zhang, Xiaohan Zhang, Yifan Wang, and Kai Gao. 2024. Token-Level Contrastive Learning with Modality-Aware Prompting for Multimodal Intent Recognition. InProceedings of the 37th AAAI Conference on Artific","arxiv_id":"2607.21908","detector":"doi_compliance","evidence":{"ref_index":49,"verdict_class":"incontrovertible","resolved_title":"Token-Level Contrastive Learning with Modality-Aware Prompting for Multimodal Intent Recognition","printed_excerpt":"10.1609/aaai","reconstructed_doi":"10.1609/aaai.v38i15.29656"},"severity":"advisory","ref_index":49,"audited_at":"2026-08-01T06:38:42.552306Z","event_type":"pith.integrity.v1","detected_doi":"10.1609/aaai.v38i15.29656","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"73347d5827c759c97e7755d9d64d1a16ddf81aef51c4e281cbd1e95661e50988","paper_version":1,"verdict_class":"incontrovertible","resolved_title":"Token-Level Contrastive Learning with Modality-Aware Prompting for Multimodal Intent Recognition","detector_version":"1.1.0","detected_arxiv_id":null,"integrity_event_id":15285,"payload_sha256":"49f09aca3946baf23586f6e65bdb6a7469c3485e21cf60137c54d3bf9fb8d2fa","signature_b64":"ypFO+iOW6+v88uQo2diTG0kYmAcKc7XBOAnHhlnNRD/hcGySJgviPFsgGxqWWPJcqFHHzg2rIn25fezRrzcUBA==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-01T06:43:32Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DmKsy4Va4WrK373M1otCeno16cO0fWAN25qHb1kZgGAjRr2UOvW3Q8BW1vTF2bhCU8JAR6h5wXX7cYk74mBRAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T16:28:10.010386Z"},"content_sha256":"c6a11018990f1345d579dade46fe46e1246707329ffc204fd97a77487d59aa57","schema_version":"1.0","event_id":"sha256:c6a11018990f1345d579dade46fe46e1246707329ffc204fd97a77487d59aa57"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JKO5OMSN4GATCWCP2E5ZWD3SGT/bundle.json","state_url":"https://pith.science/pith/JKO5OMSN4GATCWCP2E5ZWD3SGT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JKO5OMSN4GATCWCP2E5ZWD3SGT/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-05T16:28:10Z","links":{"resolver":"https://pith.science/pith/JKO5OMSN4GATCWCP2E5ZWD3SGT","bundle":"https://pith.science/pith/JKO5OMSN4GATCWCP2E5ZWD3SGT/bundle.json","state":"https://pith.science/pith/JKO5OMSN4GATCWCP2E5ZWD3SGT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JKO5OMSN4GATCWCP2E5ZWD3SGT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JKO5OMSN4GATCWCP2E5ZWD3SGT","merge_version":"pith-open-graph-merge-v1","event_count":7,"valid_event_count":7,"invalid_event_count":0,"equivocation_count":1,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"5f941621e9f5ddb319fface006b824162a93138dca0eed45109f98a9c5be0c35","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2026-07-24T02:25:27Z","title_canon_sha256":"53cbd5422ba2adecab91b521efc9593ef3da887d412acdd0d9d8bb2219109a46"},"schema_version":"1.0","source":{"id":"2607.21908","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.21908","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"arxiv_version","alias_value":"2607.21908v1","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.21908","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"pith_short_12","alias_value":"JKO5OMSN4GAT","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"pith_short_16","alias_value":"JKO5OMSN4GATCWCP","created_at":"2026-07-27T00:20:27Z"},{"alias_kind":"pith_short_8","alias_value":"JKO5OMSN","created_at":"2026-07-27T00:20:27Z"}],"graph_snapshots":[{"event_id":"sha256:ec3daf69311eb490607776cece4c6f181c9121347a319b2257e784c10db2c258","target":"graph","created_at":"2026-07-27T00:20:27Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2607.21908/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Unsupervised multimodal intent discovery aims to uncover latent intents from unlabeled multimodal dialogues, but remains challenging due to the lack of explicit semantic supervision. Existing methods often provide limited interpretability, as their refinement mainly relies on geometric similarity rather than high-level semantic guidance. To address these limitations, we propose MCSP, a fully unsupervised method that introduces semantic refinement based on concepts into multimodal intent discovery. To obtain reliable semantic evidence for intent discovery, we identify high-quality representativ","authors_text":"Hua Xu, Qianrui Zhou, Yunjin Gu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2026-07-24T02:25:27Z","title":"Unsupervised Multimodal Intent Discovery via MLLM-Guided Concept Generation and Semantic Propagation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.21908","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:92fb45a789017be9fd47cdeba7f977178b2537ea1a2b12a33384abbc95ed04d4","target":"record","created_at":"2026-07-27T00:20:27Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"5f941621e9f5ddb319fface006b824162a93138dca0eed45109f98a9c5be0c35","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2026-07-24T02:25:27Z","title_canon_sha256":"53cbd5422ba2adecab91b521efc9593ef3da887d412acdd0d9d8bb2219109a46"},"schema_version":"1.0","source":{"id":"2607.21908","kind":"arxiv","version":1}},"canonical_sha256":"4a9dd7324de18131584fd13b9b0f7234c872e96f24e176e0d6f7340df5a14f21","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4a9dd7324de18131584fd13b9b0f7234c872e96f24e176e0d6f7340df5a14f21","first_computed_at":"2026-07-27T00:20:27.437480Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-27T00:20:27.437480Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zY3/vMPh9cPqEJ7ofVq0h+GZFPPJ3vJShseVWm4brTXRCpw19EmX9d9IvUAry6NykRsw7DRPDLJ1E6zqIiOfDg==","signature_status":"signed_v1","signed_at":"2026-07-27T00:20:27.438292Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.21908","source_kind":"arxiv","source_version":1}}},"equivocations":[{"signer_id":"pith.science","event_type":"integrity_finding","target":"integrity","event_ids":["sha256:0fa579f523a2d9879cdfd512046245f357ecd4c83012dd202668c8c856a534c9","sha256:3ce8204dfb074979918f40ca4977c2b15632472df1c4acded557d1a7b44ac48c","sha256:6972f35cf3432a5b7493182b267029224db4d6b93000b086ec1e350245a5db90","sha256:c6a11018990f1345d579dade46fe46e1246707329ffc204fd97a77487d59aa57","sha256:ccf3a5c34bf8d74019ecfcb934bdf0bdbdc2c725ebad2ebd5441e438811cd54f"]}],"invalid_events":[],"applied_event_ids":["sha256:92fb45a789017be9fd47cdeba7f977178b2537ea1a2b12a33384abbc95ed04d4","sha256:ec3daf69311eb490607776cece4c6f181c9121347a319b2257e784c10db2c258"],"state_sha256":"ab3912545797e6fa64377d7265bc8e25c6d40003ca17ef81e34c58e1e1d554dd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IVLyn1zneHhhcSqw7tvvO688FCk0FoSs4jzQRVS73yQfabUwRLj7jHobasrgk3JVhUagyQEae8IVafDKLetaCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T16:28:10.013827Z","bundle_sha256":"5ea557330c00e164d2d8911b0b72e0adaecbfd613cd2ce81160bc59f2c9e0fda"}}