{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.05761", "kind": "paper", "url": "https://arxiv.org/abs/2606.05761"}, {"id": "gh:yummytanmo/subtlememory", "kind": "repo", "url": "https://github.com/Yummytanmo/SubtleMemory"}], "authors": ["Wenxuan Wang", "Haoyu Sun", "Fukuan Hou", "Mingyang Song", "Weinan Zhang", "Yu Cheng", "Yang Yang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.05761", "source_url": "https://arxiv.org/abs/2606.05761"}, "categories": ["General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "SubtleMemory evaluates fine-grained relational memory discrimination in long-horizon AI agents. It contains 1,522 evaluation instances over 10 long histories, grounded in 1,090 relation-controlled memory-variant sets, spanning user-related and non-user-related queries."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.05761", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.05761", "title": "SubtleMemory"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2606.05761", "modality": null, "name": "SubtleMemory", "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_subtlememory_d858f4b9", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "106c20cc64dae3a2d270fd46bf9fcb685f20cfa72f9c3081a9280decfaa57da0", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.05761"}, "publisher": null, "released": "2026-06-04", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.05761", "source_url": "https://arxiv.org/abs/2606.05761"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-05761", "source": "claire_radar", "source_benchmark_id": "2606.05761", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["General AI"], "area": "Language & Knowledge", "attention": {"asOf": "2026-10-05", "attemptedAt": "2026-10-04T15:31:48.030916Z", "githubRepo": "https://github.com/Yummytanmo/SubtleMemory", "githubScope": "benchmark_repo", "githubStars": 21, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": 20, "hfPaperUrl": "https://huggingface.co/papers/2606.05761", "observedAt": "2026-10-04T15:31:48.030916Z", "signalStatus": {"githubStars": {"observedAt": "2026-10-04T15:31:48.030916Z", "state": "fresh"}, "hfDatasetDownloads": {"state": "not_applicable"}, "hfDatasetLikes": {"state": "not_applicable"}, "hfPaperUpvotes": {"checkedAt": "2026-10-04T15:31:48.030916Z", "lastSuccessfulDate": "2026-09-25", "state": "stale"}}, "status": "fresh"}, "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": "c60f79137f82b8c4cb4a9959e7c1b3b6742fc91694d53915bf360fb02c37b4a8", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "SubtleMemory evaluates fine-grained relational memory discrimination in long-horizon AI agents. It contains 1,522 evaluation instances over 10 long histories, grounded in 1,090 relation-controlled memory-variant sets, spanning user-related and non-user-related queries.", "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": "To address this gap, we introduce SubtleMemory, a benchmark for fine-grained relational memory discrimination in long-running AI agents."}, "familyId": "bmf_1e2fb9ee2660", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [{"evidence": [{"basis": "existing-code-link", "linkedUrl": "https://github.com/Yummytanmo/SubtleMemory", "url": "https://github.com/Yummytanmo/SubtleMemory"}, {"basis": "existing-metric-association", "linkedUrl": "https://github.com/Yummytanmo/SubtleMemory", "url": "https://huggingface.co/papers/2606.05761"}], "scope": "benchmark_repo", "url": "https://github.com/Yummytanmo/SubtleMemory", "verification": "source-linked"}], "status": "linked"}, "heat": null, "id": "bm_subtlememory_d858f4b9", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": "https://github.com/Yummytanmo/SubtleMemory", "data": null, "hfPaper": "https://huggingface.co/papers/2606.05761", "paper": "https://arxiv.org/abs/2606.05761", "pdf": "https://arxiv.org/pdf/2606.05761", "project": null, "report": "https://arxiv.org/abs/2606.05761"}, "metrics": [], "motivation": "Persistent AI assistants, such as OpenClaw, accumulate large collections of related memories over long-term interactions.", "name": "SubtleMemory", "oneLine": "SubtleMemory evaluates fine-grained relational memory discrimination in long-horizon AI agents. It contains 1,522 evaluation instances over 10 long histories, grounded in 1,090 relation-controlled memory-variant sets, spanning user-related and non-user-related queries.", "paperTitle": "SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.05761", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "SubtleMemory Project", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://github.com/Yummytanmo/SubtleMemory"}], "ranking": {}, "readiness": "Runnable", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-04", "source": {"authors": ["Wenxuan Wang", "Haoyu Sun", "Fukuan Hou", "Mingyang Song", "Weinan Zhang", "Yu Cheng", "Yang Yang"], "categories": ["cs.AI", "cs.CL"], "id": "2606.05761", "title": "SubtleMemory: A Benchmark for Fine-Grained Relational Memory Discrimination in Long-Horizon AI Agents", "type": "arxiv", "url": "https://arxiv.org/abs/2606.05761"}, "sourceUpdatedAt": "2026-06-05T04:28:58Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Existing long-term memory benchmarks rarely probe how agents preserve and use relations among memories during downstream tasks. SubtleMemory provides a relation-controlled evaluation to measure this capability."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
