{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.26654", "kind": "paper", "url": "https://arxiv.org/abs/2606.26654"}], "authors": ["Qinkai Zhang", "Yanyan Zhao", "Xin Lu", "Yulin Hu", "Pengtao Han", "Bing Qin"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.26654", "source_url": "https://arxiv.org/abs/2606.26654"}, "categories": ["General AI", "Multimodal"], "description": {"en": "SocialPersona evaluates multimodal LLMs on recovering revealed preferences from longitudinal social-media timelines and using them in dialogue. It includes 171 user timelines, 2,597 human-verified preference tags across seven domains, and supports profile construction and response generation tasks."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.26654", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.26654", "title": "SocialPersona"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.26654", "modality": null, "name": "SocialPersona", "openness": {"code_license": null, "data_license": null, "evidence": [], "status": "unknown"}, "provenance": {"confidence": "High", "crawl_bundle": "claire_radar_2026-10-14", "crawled_at": "2026-10-14T09:44:51Z", "data_status": "primary-source-indexed", "display_eligible": "unknown", "origin_record_id": "bm_socialpersona_4370a8a3", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "7b4207fb217375d76098c69f18a7325aa8da3223c02da9fc231f7a1a86a64cde", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.26654"}, "publisher": null, "released": "2026-06-25", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.26654", "source_url": "https://arxiv.org/abs/2606.26654"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-26654", "source": "claire_radar", "source_benchmark_id": "2606.26654", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Multimodal", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 0, "hfPaperUrl": "https://huggingface.co/papers/2606.26654", "observedAt": "2026-10-04T03:02:42.754940Z", "signalStatus": {"githubStars": {"state": "not_applicable"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-10-04", "state": "stale"}}, "status": "stale"}, "capabilities": [], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "47008b2a46569b0f3b0a4421f92d644c779fa1404b9c51fb27ca0fba3eb377bb", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "SocialPersona evaluates multimodal LLMs on recovering revealed preferences from longitudinal social-media timelines and using them in dialogue. It includes 171 user timelines, 2,597 human-verified preference tags across seven domains, and supports profile construction and response generation tasks.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We introduce SocialPersona, a benchmark for evaluating whether multimodal large language models (MLLMs) can recover revealed preferences from longitudinal social-media timelines and use them in dialogue."}, "familyId": "bmf_6f5f69e2ce5e", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_socialpersona_4370a8a3", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.26654", "paper": "https://arxiv.org/abs/2606.26654", "pdf": "https://arxiv.org/pdf/2606.26654", "project": null, "report": "https://arxiv.org/abs/2606.26654"}, "metrics": [], "motivation": "Personalized language-model assistants are often evaluated through a memory lens: can a model recall preferences users have explicitly stated in dialogue?", "name": "SocialPersona", "oneLine": "SocialPersona evaluates multimodal LLMs on recovering revealed preferences from longitudinal social-media timelines and using them in dialogue. It includes 171 user timelines, 2,597 human-verified preference tags across seven domains, and supports profile construction and response generation tasks.", "paperTitle": "SocialPersona: Benchmarking Personalized Profiling and Response with Multimodal Social-Media Context", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.26654", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "Hugging Face", "organizationType": "benchmark-organization", "role": "benchmark-publisher", "sourceUrl": "https://huggingface.co/papers/2606.26654"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-25", "source": {"authors": ["Qinkai Zhang", "Yanyan Zhao", "Xin Lu", "Yulin Hu", "Pengtao Han", "Bing Qin"], "categories": ["cs.CL", "cs.HC", "cs.IR"], "id": "2606.26654", "title": "SocialPersona: Benchmarking Personalized Profiling and Response with Multimodal Social-Media Context", "type": "arxiv", "url": "https://arxiv.org/abs/2606.26654"}, "sourceUpdatedAt": "2026-06-25T06:31:56Z", "topics": ["Multimodal"], "venueAttempts": [], "whyItMatters": "Existing personalization benchmarks rely on explicitly stated preferences; SocialPersona tests inference from natural multimodal traces, a harder capability. It provides a reusable benchmark for measuring progress on long-horizon user modeling and personalized dialogue."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
