Provider and attribution
LMArena Leaderboard Dataset
- Registry state
- registered
- Rights state
- approved
- Provider group
- lmarena
Contains data from the Arena Leaderboard Dataset by Arena, licensed under CC BY 4.0
Release planning · no model scores
Search and document arenas evidence mapping
Exact V3 collection, facet, metric, lineage, identity and rights mapping for Search and document arenas from LMArena Leaderboard Dataset.
V3 evidence records
1 collection mapping
A repeated upstream benchmark can map to several collections. Each record below preserves its own task role, facets, capability atoms, subject identity, metrics, lineages and exclusions.
supporting evidence · research-synthesis/lmarena/search-and-document-arenas
Research & Evidence Synthesis
Which models are best at research and evidence synthesis?
- Upstream access
- machine-ready
- Eligibility
- runnable-now
- Registry
- registered
- Max task directness
- 45%
Collection facets
Evidence retrieval · hybrid · 25%
Find relevant, authoritative evidence and avoid source laundering.
Source reconciliation · hybrid · 25%
Resolve disagreement, dates, definitions, and incompatible scopes.
Competitive and market analysis · hybrid · 25%
Separate observed facts, inference, and speculation.
Research briefs · hybrid · 25%
Produce a concise, cited, decision-useful synthesis.
Capability atoms
search_preferencedocument_preference
Subject identity
- Declared:
- harnessed-model
- Metric partitions:
- foundation-model, harnessed-model
- Separate partition required:
- yes
- Identity projection:
- forbidden
Usable evidence lineages
lmarena-document-style-control/foundation-modellmarena-search-style-control/harnessed-model
Exact metric contract
| Metric | Lineage | Subject | Protocol | Use |
|---|---|---|---|---|
search_style_control.overallLMArena Search (style controlled) — Overall | lmarena-search-style-control | harnessed-model | LMArena Search Arena style-controlled Bradley-Terry rating | usable |
document_style_control.overallLMArena Document (style controlled) — Overall | lmarena-document-style-control | foundation-model | LMArena Document Arena style-controlled Bradley-Terry rating | usable |
Exclusions and blockers
- metric subject identity differs from the editorial mapping; publish and calibrate each identity partition separately
Mapping note
Human preference supports but does not prove citation correctness.