{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2606.17328", "kind": "paper", "url": "https://arxiv.org/abs/2606.17328"}], "authors": ["Xianxuan Long", "Zhikai Chen", "Shenglai Zeng", "Shouren Wang", "Kai Guo", "Jiliang Tang"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2606.17328", "source_url": "https://arxiv.org/abs/2606.17328"}, "categories": ["Factuality", "General AI", "Language & Knowledge", "cs.AI"], "description": {"en": "MemTrace evaluates long-term memory in LLM agents at the knowledge point level, probing memory age, question type, and evidence condition. Scoring is based on accuracy across these dimensions."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2606.17328", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2606.17328", "title": "MemTrace"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "institutions": ["Michigan State University", "Case Western Reserve University"], "institutions_reference": {"basis": "openalex_linked_institutions", "doi": "10.48550/arxiv.2606.17328", "retrieved_at": "2026-11-04", "source_key": "claire-radar:2606.17328", "source_url": "https://openalex.org/W7165062880"}, "key": "claire-radar:2606.17328", "modality": null, "name": "MemTrace", "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_memtrace_56029435", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "cf244536515190edcfef21bad9d78b1253c9cf3228aca5f00bd427a205aef0e9", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2606.17328"}, "publisher": null, "released": "2026-06-15", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2606.17328", "source_url": "https://arxiv.org/abs/2606.17328"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2606-17328", "source": "claire_radar", "source_benchmark_id": "2606.17328", "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": null, "githubScope": null, "githubStars": null, "hfDailySubmittedAt": null, "hfDataset": null, "hfDatasetDownloads": null, "hfDatasetLikes": null, "hfPaperUpvotes": null, "hfPaperUrl": "https://huggingface.co/papers/2606.17328", "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", "state": "unavailable"}}, "status": "unavailable"}, "capabilities": ["Factuality"], "confidence": "High", "construction": "Unknown", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-21T04:30:40.934319Z", "inputHash": "337defe4b68c90141b6c7e584e0aeee3beabd9ddf9ed8399eaa29a648260620d", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "MemTrace evaluates long-term memory in LLM agents at the knowledge point level, probing memory age, question type, and evidence condition. Scoring is based on accuracy across these dimensions.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["exact coined title identity tied to benchmark evidence", "benchmark release stated in one sentence"], "snippet": "We introduce MemTrace, a benchmark whose unit of measurement is the knowledge point: a single typed fact about the user, rather than an individual question."}, "familyId": "bmf_cc4e1edaadda", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_memtrace_56029435", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2606.17328", "paper": "https://arxiv.org/abs/2606.17328", "pdf": "https://arxiv.org/pdf/2606.17328", "project": null, "report": "https://arxiv.org/abs/2606.17328"}, "metrics": [], "motivation": "LLM agents increasingly maintain long-term memory of user facts across sessions.", "name": "MemTrace", "oneLine": "MemTrace evaluates long-term memory in LLM agents at the knowledge point level, probing memory age, question type, and evidence condition. Scoring is based on accuracy across these dimensions.", "paperTitle": "MemTrace: Probing What Final Accuracy Misses in Long-Term Memory", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2606.17328", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "arXiv", "organizationType": "community", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2606.17328"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-15", "source": {"authors": ["Xianxuan Long", "Zhikai Chen", "Shenglai Zeng", "Shouren Wang", "Kai Guo", "Jiliang Tang"], "categories": ["cs.AI"], "id": "2606.17328", "title": "MemTrace: Probing What Final Accuracy Misses in Long-Term Memory", "type": "arxiv", "url": "https://arxiv.org/abs/2606.17328"}, "sourceUpdatedAt": "2026-06-15T22:21:23Z", "topics": ["cs.AI"], "venueAttempts": [], "whyItMatters": "Aggregated accuracy misses important memory behaviors. MemTrace provides finer-grained metrics to reveal bottlenecks in evidence use, guiding improvements in memory systems."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
