Update model card metadata and benchmark provenance

README.md

@@ -60,7 +60,7 @@ Instead of generating text token by token, ClassOne evaluates structured decisio
 
 ## Official Results (Benchmark Heaven)
 
-Official third-party JevBench leaderboards are hosted at **[Benchmark Heaven](https://benchmarkheaven.com/jev-models)** (checked `2026-11-13`, benchmark release `v1.8.8`; see the live entry for [`classone-qwen3.5-9b`](https://benchmarkheaven.com/jev-models/classone-qwen3.5-9b)). Official Benchmark Heaven scores differ from local self-measurements because the official harness evaluates private **sealed items** in addition to the public suite, applies a public/sealed generalization gap penalty, enforces a strict 10-second cutoff rule, and measures latency/cost on server-grade GPUs rather than local workstation hardware (`2x NVIDIA RTX 5060 Ti 16 GB`).
+Official third-party JevBench leaderboards are hosted at **[Benchmark Heaven](https://benchmarkheaven.com/jev-models)** (checked `2026-11-13`, JevBench `v1.8.10` / method `v1.7`, published `2026-11-12`; see the live entry for [`classone-qwen3.5-9b`](https://benchmarkheaven.com/jev-models/classone-qwen3.5-9b)). Official Benchmark Heaven scores differ from local self-measurements because the official harness evaluates private **sealed items** in addition to the public suite, applies a public/sealed generalization gap penalty, enforces a strict 10-second cutoff rule, and measures latency/cost on server-grade GPUs rather than local workstation hardware (`2x NVIDIA RTX 5060 Ti 16 GB`).
 
 ## Edge vs Cloud Latency (Self-Measured Comparison)
 
@@ -77,7 +77,7 @@ Measured against a remote cloud LLM decision API round-trip:
 - **`Choice`** — Categorical selection over 2–255 dynamic options with full probability distribution
 - **`Score`** — Continuous ordinal rubric rating over 2–10 levels (expected value)
 
-All outputs are calibrated with a combined NLL + normalized Brier loss and post-hoc temperature scaling.
+All outputs are calibrated with a combined focal NLL (γ=2.0) + normalized Brier + contrastive margin loss and post-hoc temperature scaling.
 
 ## Quickstart