{"record": {"aliases": [], "artifacts": [{"id": "arxiv:2608.07540", "kind": "paper", "url": "https://arxiv.org/abs/2608.07540"}], "authors": ["Fateme Mazdarani", "Carlos Toxtli"], "authors_reference": {"basis": "arxiv_metadata", "retrieved_at": "2026-11-03", "source_key": "claire-radar:2608.07540", "source_url": "https://arxiv.org/abs/2608.07540"}, "categories": ["Factuality", "Language & Knowledge", "Mathematics & Formal Science", "cs.AI"], "description": {"en": "TREAT evaluates LLMs' ability to recognize theorem identities from equivalence-preserving formula transformations, with 737 identities and 29,480 transformed rows."}, "documents": [{"document_type": "registry_page", "id": "claire_radar:https://arxiv.org/abs/2608.07540", "source": "claire_radar", "source_url": "https://arxiv.org/abs/2608.07540", "title": "TREAT"}], "evidence_summary": {"document_count": 1, "model_count": null, "model_count_basis": null}, "key": "claire-radar:2608.07540", "modality": null, "name": "TREAT", "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_treat_ece84fbf", "origin_source": "arxiv", "recognition_confidence": "1.0", "record_sha256": "0b5c31fa91c05fd9def24ec4c4fdb602a886e32b2f5ea1b26404d8e5dc2e66d7", "relation": "introduces", "review_state": "unreviewed", "source_url": "https://arxiv.org/abs/2608.07540"}, "publisher": null, "released": "2026-07-29", "released_reference": {"basis": "benchmark_release", "source_key": "claire-radar:2608.07540", "source_url": "https://arxiv.org/abs/2608.07540"}, "schema_version": 1, "sizes": [], "slug": "claire-radar-2608-07540", "source": "claire_radar", "source_benchmark_id": "2608.07540", "source_metadata": {"claire_radar": {"adoption30d": null, "aliases": [], "annotation": "Unknown", "applicationDomains": ["Mathematics & Formal Science"], "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/2608.07540", "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": "Transform Existing", "constructionDetail": "Unknown — the indexer does not infer construction details without explicit source evidence.", "copyGeneration": {"generatedAt": "2026-08-20T15:22:10.044731Z", "inputHash": "858fffa6bbbcf8b0a17836ac548bed4a1083e14912b97c353b2448955ad8e95e", "model": "deepseek-v4-flash"}, "dataStatus": "primary-source-indexed", "demo": false, "description": "TREAT evaluates LLMs' ability to recognize theorem identities from equivalence-preserving formula transformations, with 737 identities and 29,480 transformed rows.", "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 TREAT, a benchmark for evaluating whether large language models can recover known theorem identities from equivalence-preserving formula-level transformations."}, "familyId": "bmf_a5075550e1f2", "firstSeenAt": "2026-08-19", "githubIndex": {"candidateCount": 0, "indexedAt": "2026-09-17T04:40:34.111665+00:00", "repositories": [], "status": "not-found"}, "heat": null, "id": "bm_treat_ece84fbf", "indexedAt": "2026-08-19T11:05:13.395391Z", "industrySectors": [], "links": {"code": null, "data": null, "hfPaper": "https://huggingface.co/papers/2608.07540", "paper": "https://arxiv.org/abs/2608.07540", "pdf": "https://arxiv.org/pdf/2608.07540", "project": null, "report": "https://arxiv.org/abs/2608.07540"}, "metrics": [], "motivation": "AI systems increasingly operate between flexible input representations and formal objects used by downstream tools.", "name": "TREAT", "oneLine": "TREAT evaluates LLMs' ability to recognize theorem identities from equivalence-preserving formula transformations, with 737 identities and 29,480 transformed rows.", "paperTitle": "TREAT: Evaluating Access to Formal Knowledge across Equivalent Mathematical Representations", "primaryDomain": "Mathematics & Formal Science", "publication": {"evidence": "Accepted at 28th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC 2026)", "evidenceLevel": "author-claim", "evidenceUrl": "https://arxiv.org/abs/2608.07540", "source": "arxiv-comments", "status": "acceptance_claimed", "venue": "28th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC 2026)", "verifiedAt": "2026-10-14T01:46:58.058199Z"}, "publications": [], "ranking": {"90d": {"confidence": "Low", "coverage": 0.0, "growth": {"components": {"githubStars": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfDatasetDownloads": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}, "hfPaperUpvotes": {"normalization": "percentile", "normalized": null, "percentile": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "elapsedDays": null, "historyDate": null, "rank": null, "score": null, "windowDays": 90}, "level": {"components": {"githubStars": {"normalization": "log-max", "normalized": null, "value": null}, "hfDatasetDownloads": {"normalization": "log-max", "normalized": null, "value": null}, "hfPaperUpvotes": {"normalization": "log-max", "normalized": null, "value": null}}, "confidence": "Low", "coverage": 0.0, "rank": null, "score": null}, "method": "attention-ranking-v13", "rank": null, "score": null}}, "readiness": "Paper only", "recognitionConfidence": 1.0, "relation": "introduces", "releasedAt": "2026-07-29", "source": {"authors": ["Fateme Mazdarani", "and Carlos Toxtli"], "categories": ["cs.AI"], "id": "2608.07540", "title": "TREAT: Evaluating Access to Formal Knowledge across Equivalent Mathematical Representations", "type": "arxiv", "url": "https://arxiv.org/abs/2608.07540"}, "sourceUpdatedAt": "2026-07-29T23:02:46Z", "topics": ["cs.AI"], "venueAttempts": [{"decisionRaw": "Accepted at 28th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC 2026)", "evidence": [{"level": "author-claim", "observedAt": "2026-10-14T01:46:58.058199Z", "rawValue": "Accepted at 28th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC 2026)", "sourceType": "arxiv-comments", "sourceUrl": "https://arxiv.org/abs/2608.07540"}], "reviewStatus": "accepted", "venueName": "28th International Symposium on Symbolic and Numeric Algorithms for Scientific Computing (SYNASC 2026)"}], "whyItMatters": "It targets representation-robust access to formal knowledge, critical for AI tools interacting with formal systems."}}}, "schema_version": 1, "scores_by_source": {}, "siblings": []}
