{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:SWD46MLNIKJFFMHDX32FXTD37U","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"dfc95a3223dd45e944a436b38d9abe6e2adab1d10e7dfecfc3b17f3f0b1545c0","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-16T04:22:47Z","title_canon_sha256":"1e2c8be8716ead75032c457bcef43681811d7b1966927d23f18e0f4fd6252a88"},"schema_version":"1.0","source":{"id":"2605.16806","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.16806","created_at":"2026-05-20T00:03:23Z"},{"alias_kind":"arxiv_version","alias_value":"2605.16806v1","created_at":"2026-05-20T00:03:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.16806","created_at":"2026-05-20T00:03:23Z"},{"alias_kind":"pith_short_12","alias_value":"SWD46MLNIKJF","created_at":"2026-05-20T00:03:23Z"},{"alias_kind":"pith_short_16","alias_value":"SWD46MLNIKJFFMHD","created_at":"2026-05-20T00:03:23Z"},{"alias_kind":"pith_short_8","alias_value":"SWD46MLN","created_at":"2026-05-20T00:03:23Z"}],"graph_snapshots":[{"event_id":"sha256:680ac0c82adbba216c527f290beec7e07edd8f79d9ecf02e23d0c23ef0f26b3b","target":"graph","created_at":"2026-05-20T00:03:23Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"The CAMA module explicitly models inter-modal relationships via affinity matrices and enforces cross-modal consistency through contrastive learning, enabling adaptive suppression of uninformative modalities without discarding them and yielding consistent improvements over unimodal baselines and prior cross-attention approaches."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That affinity matrices derived from the heterogeneous features (facial action units, head pose, eye gaze, interaction logs) plus contrastive learning will reliably capture and mitigate modality degradation in the specific 50-student EcoJourneys dataset without introducing artifacts that affect downstream satisfaction prediction."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"AAMLA aligns multimodal student data using affinity matrices and contrastive learning to predict collaboration satisfaction in game-based learning despite modality degradation."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A module using affinity matrices and contrastive learning to align and selectively suppress unreliable data sources improves predictions of student collaboration satisfaction in game-based learning."}],"snapshot_sha256":"315256b5cc5a29abb40074ffb9087dd4ce1541b7ea396269f5fe3f50bd39a96a"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T21:31:19.287363Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T21:21:18.427403Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T19:01:56.282518Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T18:33:26.420046Z","status":"skipped","version":"1.0.0"}],"endpoint":"/pith/2605.16806/integrity.json","findings":[],"snapshot_sha256":"9ffed6c367bcbbfbc6f76ed1fc99d605c4f66f0239f2873075f75d04b206a72b","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Collaborative game-based learning environments offer rich opportunities for small-group knowledge construction, yet automatically predicting student collaboration satisfaction remains challenging. A critical barrier is modality degradation: in educational deployments, individual modalities such as eye gaze exhibit inconsistent informativeness across student cohorts, causing implicit attention-based fusion to produce brittle multimodal representations. We propose the Affinity-Aligned Multimodal Learning Analytics (AAMLA) framework, whose core contribution is the Cross-modal Affinity-guided Moda","authors_text":"Chia-Ming Lee, Wen-Hsin Tsai, Yuk-Ying Tung","cross_cats":["cs.AI","cs.CV"],"headline":"A module using affinity matrices and contrastive learning to align and selectively suppress unreliable data sources improves predictions of student collaboration satisfaction in game-based learning.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-16T04:22:47Z","title":"Cross-modal Affinity-aligned Multimodal Learning Analytics for Predicting Student Collaboration Satisfaction in Game-Based Learning"},"references":{"count":52,"internal_anchors":0,"resolved_work":52,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Halim Acosta, Seung Lee, Bradford Mott, Haesol Bae, Krista Glazewski, Cindy Hmelo-Silver, and James C. Lester. Multimodal learning analytics for predicting student collab- oration satisfaction in coll","work_id":"b553c0da-2469-4026-9a9f-1bef8c52c5d0","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Waleed Mugahed Al-Rahmi and Mohd Shahizan Othman. Evaluating student’s satisfaction of using social media through collaborative learning in higher education.Interna- tional Journal of Advances in Engi","work_id":"51c28476-719d-48f0-b0cd-5080e170a5eb","year":2013},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Comparing collabo- rative and cooperative gameplay for academic and gaming achievements.Journal of Educational Computing Research, 57(8):2110–2140, 2020","work_id":"23e7b973-ac5b-4b06-8e85-47cf659ca7a4","year":2020},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Openface 2.0: Facial behavior analysis toolkit","work_id":"5131a68d-e4ed-4504-b011-db0f898013f3","year":null},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Classifying confusion: Autodetec- tion of communicative misunderstandings using facial action units","work_id":"12f66954-d273-4596-90e6-5b3897d5d129","year":2019}],"snapshot_sha256":"1dc08ed990eebeacef10137aba570aa07155d0eabaa4c4eb51c435c098e8203f"},"source":{"id":"2605.16806","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-19T21:09:53.457556Z","id":"52bcc7d5-2556-4319-8559-356f23db5352","model_set":{"reader":"grok-4.3"},"one_line_summary":"AAMLA aligns multimodal student data using affinity matrices and contrastive learning to predict collaboration satisfaction in game-based learning despite modality degradation.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A module using affinity matrices and contrastive learning to align and selectively suppress unreliable data sources improves predictions of student collaboration satisfaction in game-based learning.","strongest_claim":"The CAMA module explicitly models inter-modal relationships via affinity matrices and enforces cross-modal consistency through contrastive learning, enabling adaptive suppression of uninformative modalities without discarding them and yielding consistent improvements over unimodal baselines and prior cross-attention approaches.","weakest_assumption":"That affinity matrices derived from the heterogeneous features (facial action units, head pose, eye gaze, interaction logs) plus contrastive learning will reliably capture and mitigate modality degradation in the specific 50-student EcoJourneys dataset without introducing artifacts that affect downstream satisfaction prediction."}},"verdict_id":"52bcc7d5-2556-4319-8559-356f23db5352"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:485a94a366e21776f0fca7d8e3e52f62c965709f9d65702574ea822c646321d8","target":"record","created_at":"2026-05-20T00:03:23Z","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":"dfc95a3223dd45e944a436b38d9abe6e2adab1d10e7dfecfc3b17f3f0b1545c0","cross_cats_sorted":["cs.AI","cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-16T04:22:47Z","title_canon_sha256":"1e2c8be8716ead75032c457bcef43681811d7b1966927d23f18e0f4fd6252a88"},"schema_version":"1.0","source":{"id":"2605.16806","kind":"arxiv","version":1}},"canonical_sha256":"9587cf316d429252b0e3bef45bcc7bfd01971e56a569474d196725599cba2b9f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9587cf316d429252b0e3bef45bcc7bfd01971e56a569474d196725599cba2b9f","first_computed_at":"2026-05-20T00:03:23.194458Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:03:23.194458Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kjzhoeoxzu9ISsDmaMs3+EhRyOmxgM7yk68YzudHfr0tMovm9KMkgDY4QLonXSvuMejS8ly0al6FBgIz8yfHBg==","signature_status":"signed_v1","signed_at":"2026-05-20T00:03:23.195378Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.16806","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:485a94a366e21776f0fca7d8e3e52f62c965709f9d65702574ea822c646321d8","sha256:680ac0c82adbba216c527f290beec7e07edd8f79d9ecf02e23d0c23ef0f26b3b"],"state_sha256":"063c4d341ce9fa267fe388db4da289dd08fcbf25be1c65749cc8f0380c93f812"}