Align user-facing labels with paper terminology

Sweeps every visible "Composite benchmark" / "Merged Benchmark" /
"Independently evaluated" / "Documentation gaps" label so the UI
uses the rollout-hierarchy and interpretive-signal vocabulary from
the paper (§3.2 and §4.2.1).

- "Composite benchmark" -> "Suite" (paper §3.2)
- "Merged Benchmark" -> "Suite"
- "N composite benchmarks" -> "N component benchmarks"
- "average of N composite scores" -> "average of N component scores"
- "Independently evaluated" -> "Third-party reported" (paper §4.2.1)
- "Documentation gaps" -> "Reproducibility gap"
- Provenance tooltips harmonized across both badge sites:
first-party / third-party / collaborative wording matches paper
- "Reading mode" / "Research-first reading mode" -> "Reader mode" /
"Research reader mode" (paper §4.2.2)

No backend or data-shape changes; internal fields like
composite_benchmark_name are unchanged.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

app/evals/[id]/page.tsx

@@ -167,7 +167,7 @@ export default function EvalDetailPage() {
             </Button>
             <div className="text-center">
               <h2 className="text-xl font-medium tracking-tight text-foreground/90 md:text-2xl">
-                {isComposite ? "Composite benchmark" : "Single benchmark details"}
+                {isComposite ? "Suite" : "Single benchmark details"}
               </h2>
             </div>
             <div />
@@ -178,7 +178,7 @@ export default function EvalDetailPage() {
             </Button>
             <div className="flex-1 text-center">
               <h2 className="text-base font-medium tracking-tight text-foreground/90">
-                {isComposite ? "Composite benchmark" : "Single benchmark details"}
+                {isComposite ? "Suite" : "Single benchmark details"}
               </h2>
             </div>
           </div>
@@ -202,7 +202,7 @@ export default function EvalDetailPage() {
 }
 
 // ---------------------------------------------------------------------------
-// Composite benchmark view
+// Suite view (paper §3.2 — composite reporting unit)
 // ---------------------------------------------------------------------------
 
 function CompositeEvalView({
@@ -225,7 +225,7 @@ function CompositeEvalView({
         <CardContent className="p-5 sm:p-6 space-y-4">
           <div className="flex flex-wrap items-center gap-2">
             <Badge variant="outline" className="text-[11px] uppercase tracking-[0.18em]">
-              Composite Benchmark
+              Suite
             </Badge>
             <Badge variant="secondary">
               {summary.aggregate_sources?.length ?? 0} metrics
@@ -241,7 +241,7 @@ function CompositeEvalView({
             </h1>
             <p className="mt-2 max-w-3xl text-sm leading-6 text-muted-foreground">
               {summary.benchmark_card?.purpose_and_intended_users?.goal
-                ?? `Composite benchmark suite with ${summary.aggregate_sources?.length ?? 0} metrics evaluated across ${summary.models_count.toLocaleString()} models.`}
+                ?? `Suite aggregating ${summary.aggregate_sources?.length ?? 0} metrics across ${summary.models_count.toLocaleString()} models.`}
             </p>
           </div>
         </CardContent>

components/benchmark-detail.tsx

@@ -2859,11 +2859,11 @@ export function BenchmarkDetail({
                     className={`inline-flex items-center rounded-full border px-2 py-0.5 text-[10px] font-semibold ${getRelationshipBadgeTone(relationship)}`}
                     title={
                       relationship === "first_party"
-                        ? "Self-reported by the model's developer."
+                        ? "Reported by the model's developer (first-party)."
                         : relationship === "third_party"
-                          ? "Independently evaluated by an outside party."
+                          ? "Reported by an independent third party."
                           : relationship === "collaborative"
-                            ? "Joint evaluation by the developer and an outside party."
+                            ? "Collaborative report by the developer and a third party."
                             : undefined
                     }
                   >
@@ -5060,11 +5060,11 @@ function BenchmarkDeepDiveDialogPanel({
                         className={`inline-flex items-center rounded-full border px-2 py-0.5 text-[10px] font-semibold ${getRelationshipBadgeTone(entry.relationship)}`}
                         title={
                           entry.relationship === "first_party"
-                            ? "Reported by the model's developer."
+                            ? "Reported by the model's developer (first-party)."
                             : entry.relationship === "third_party"
-                              ? "Independently evaluated by an outside party."
+                              ? "Reported by an independent third party."
                               : entry.relationship === "collaborative"
-                                ? "Joint evaluation by the developer and an outside party."
+                                ? "Collaborative report by the developer and a third party."
                                 : undefined
                         }
                       >

components/eval-card.tsx

@@ -120,7 +120,7 @@ export function EvalCard({ summary, delayMs = 0 }: EvalCardProps) {
         <div className="min-w-0">
           <div className="text-xl font-bold">{summary.evaluation_name}</div>
           <div className="mt-1 text-xs font-medium uppercase tracking-[0.18em] text-muted-foreground">
-            Composite benchmark: {summary.composite_benchmark_name}
+            Suite: {summary.composite_benchmark_name}
           </div>
           <div className="mt-1 text-sm text-muted-foreground line-clamp-2">
             {overviewText ?? summary.metric_config.evaluation_description}
@@ -131,13 +131,13 @@ export function EvalCard({ summary, delayMs = 0 }: EvalCardProps) {
           {summary.third_party_ratio > 0 && (
             <Badge className="bg-emerald-600 text-white hover:bg-emerald-600">
               <BadgeCheck className="mr-1 h-3 w-3" />
-              Independently evaluated
+              Third-party reported
             </Badge>
           )}
           {reproducibilityGapCount > 0 && (
             <Badge className="bg-amber-500 text-amber-950 hover:bg-amber-500">
               <AlertTriangle className="mr-1 h-3 w-3" />
-              Documentation gaps
+              Reproducibility gap
             </Badge>
           )}
         </div>

components/eval-detail.tsx

@@ -629,10 +629,10 @@ export function EvalDetail({ summary }: EvalDetailProps) {
     : "Not linked"
   const leaderboardDescription = isResearchView
     ? summary.is_aggregated
-      ? "Models ranked by average raw score across the contributing composite benchmarks."
+      ? "Models ranked by average raw score across the suite's component benchmarks."
       : "Models ranked by raw score for this benchmark."
     : summary.is_aggregated
-      ? "Averaged model results across the contributing composite benchmarks, with drill-down to each component score."
+      ? "Averaged model results across the suite's component benchmarks, with drill-down to each component score."
       : "Model results with benchmark context, source dataset detail, and optional instance-data links."
   const reportingCompleteness = summary.evalcards?.annotations?.reporting_completeness
   const benchmarkComparability = summary.evalcards?.annotations?.benchmark_comparability
@@ -695,15 +695,15 @@ export function EvalDetail({ summary }: EvalDetailProps) {
                 <div className="space-y-2.5">
                   <div className="flex flex-wrap items-center gap-2">
                     <Badge variant="outline" className="border-border/60 bg-background/80 text-[11px] uppercase tracking-[0.18em]">
-                      {summary.is_aggregated ? "Merged Benchmark" : "Single Benchmark"}
+                      {summary.is_aggregated ? "Suite" : "Single Benchmark"}
                     </Badge>
                     {summary.is_aggregated ? (
                       <Badge variant="secondary" className="font-normal">
-                        {summary.aggregate_sources?.length ?? 0} composite benchmarks
+                        {summary.aggregate_sources?.length ?? 0} component benchmarks
                       </Badge>
                     ) : (
                       <Badge variant="secondary" className="font-normal">
-                        Composite: {summary.composite_benchmark_name}
+                        Suite: {summary.composite_benchmark_name}
                       </Badge>
                     )}
                     <Badge variant="secondary" className="font-normal capitalize">
@@ -778,11 +778,11 @@ export function EvalDetail({ summary }: EvalDetailProps) {
                 <dl className="mt-3 grid gap-x-5 gap-y-3 text-sm sm:grid-cols-2 xl:grid-cols-5">
                   <div>
                     <dt className="text-[11px] font-semibold uppercase tracking-[0.16em] text-muted-foreground">
-                      Composite benchmark
+                      Suite
                     </dt>
                     <dd className="mt-1 break-words font-medium">
                       {summary.is_aggregated
-                        ? summary.aggregate_sources?.map((source) => source.composite_benchmark_name).join(", ") || "Multiple composite benchmarks"
+                        ? summary.aggregate_sources?.map((source) => source.composite_benchmark_name).join(", ") || "Multiple suites"
                         : summary.composite_benchmark_name}
                     </dd>
                   </div>
@@ -1160,7 +1160,7 @@ export function EvalDetail({ summary }: EvalDetailProps) {
                           <TableCell className="hidden md:table-cell">
                             <div className="text-sm text-muted-foreground capitalize">
                               {modelResult.aggregate_components && modelResult.aggregate_components.length > 1
-                                ? `average of ${modelResult.aggregate_components.length} composite scores`
+                                ? `average of ${modelResult.aggregate_components.length} component scores`
                                 : datasetName ?? "Detailed result source"}
                             </div>
                           </TableCell>

components/home-mode-label.tsx

@@ -5,5 +5,5 @@ import { useAudienceMode } from "@/components/audience-mode-provider"
 export function HomeModeLabel() {
   const { mode } = useAudienceMode()
 
-  return <span>{mode === "research" ? "Research-first reading mode" : "Policy-first reading mode"}</span>
+  return <span>{mode === "research" ? "Research reader mode" : "Policy reader mode"}</span>
 }

components/navigation.tsx

@@ -183,7 +183,7 @@ export function Navigation() {
 
               <div className="mt-3 border-t border-border/60 pt-3">
                 <div className="text-[10px] font-semibold uppercase tracking-[0.18em] text-muted-foreground">
-                  Reading mode
+                  Reader mode
                 </div>
                 <div className="mt-2 grid grid-cols-2 gap-2">
                   <button