{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.00009", "kind": "paper", "url": "https://arxiv.org/abs/2608.00009"}], "authors": ["Ahmed Cherif"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.00009", "source_url": "https://arxiv.org/abs/2608.00009"}, "categories": ["General AI", "Language & Knowledge", "cs.CL"], "description": {"en": "AgentMemBench evaluates five long-term memory management strategies for conversational AI agents across three public datasets (LoCoMo, MultiDoc2Dial, MSC), covering multi-session dialogue, document grounding, and persona-grounded chat. Scoring uses retrieval metrics (Recall@k, MRR, nDCG@k), Answer F1, LLM-judge faithfulness, memory footprint, and latency over 491 annotated question turns."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.00009", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.00009", "title": "AgentMemBench"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.00009", "modality": null, "name": "AgentMemBench", "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_agentmembench_9938dbed", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "e8af70e59dc872812731828bda92be3e998afbe8e10ee01aaebe6847543efecd", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.00009"}, "publisher": null, "released": "2026-06-16", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.00009", "source_url": "https://arxiv.org/abs/2608.00009"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-00009", "source": "claire_radar", "source_benchmark_id": "2608.00009", "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/2608.00009", "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": [], "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": "106686e4a9d95585a33c4ef011227553ef30c929aa78e88f7784e6e26e67e0b9", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "AgentMemBench evaluates five long-term memory management strategies for conversational AI agents across three public datasets (LoCoMo, MultiDoc2Dial, MSC), covering multi-session dialogue, document grounding, and persona-grounded chat. Scoring uses retrieval metrics (Recall@k, MRR, nDCG@k), Answer F1, LLM-judge faithfulness, memory footprint, and latency over 491 annotated question turns.", "domainCuration": {"method": "rules-v1", "state": "auto"}, "evidence": {"reasonCodes": ["coined title prefix ending in Bench or Benchmark", "benchmark release stated in one sentence", "evaluation protocol evidence"], "snippet": "We present AgentMemBench, a unified, reproducible benchmark evaluating five memory management strategies under identical conditions: in-context windowing (ICW), external key-value store (EKV), graph-based episodic memory (GEM), compression-based summarisation (CBS), and web-augmented memory (WAM)."}, "familyId": "bmf_9d9e93f978b2", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_agentmembench_9938dbed", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2608.00009", "paper": "https://arxiv.org/abs/2608.00009", "pdf": "https://arxiv.org/pdf/2608.00009", "project": null, "report": "https://arxiv.org/abs/2608.00009"}, "metrics": [], "motivation": "Long-term memory remains a critical bottleneck for conversational AI agents, whose finite context windows cannot support coherent recall across thousands of turns.", "name": "AgentMemBench", "oneLine": "AgentMemBench evaluates five long-term memory management strategies for conversational AI agents across three public datasets (LoCoMo, MultiDoc2Dial, MSC), covering multi-session dialogue, document grounding, and persona-grounded chat. Scoring uses retrieval metrics (Recall@k, MRR, nDCG@k), Answer F1, LLM-judge faithfulness, memory footprint, and latency over 491 annotated question turns.", "paperTitle": "AgentMemBench: A Systematic Benchmark for Evaluating Long-Term Memory Management Strategies in Conversational AI Agents", "primaryDomain": "General AI", "publication": {"evidence": null, "evidenceLevel": "unverified", "evidenceUrl": "https://arxiv.org/abs/2608.00009", "source": "arxiv-metadata", "status": "unverified", "venue": null, "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "publishers": [{"name": "AgentMemBench Team", "organizationType": "academic-lab", "role": "benchmark-publisher", "sourceUrl": "https://arxiv.org/abs/2608.00009"}], "ranking": {}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-06-16", "source": {"authors": ["Ahmed Cherif"], "categories": ["cs.CL", "cs.AI"], "id": "2608.00009", "title": "AgentMemBench: A Systematic Benchmark for Evaluating Long-Term Memory Management Strategies in Conversational AI Agents", "type": "arxiv", "url": "https://arxiv.org/abs/2608.00009"}, "sourceUpdatedAt": "2026-06-16T08:27:18Z", "topics": ["cs.CL"], "venueAttempts": [], "whyItMatters": "Long-term memory is a key bottleneck for conversational agents. AgentMemBench provides a controlled, reproducible comparison of memory strategies under identical conditions, enabling practitioners to make informed trade-offs between recall quality, accuracy, and resource cost."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
