Clarify JEVision evaluation claims and update graphics

README.md

@@ -33,9 +33,9 @@ The release combines a trained text adapter, a separately trained visual sidecar
 | Request envelope | Configured for up to 80,000 processed input tokens |
 | Packaging | Adapters, heads, and a runnable local server; base weights download separately |
 
-![Jev, KEV-0.8B and JEVision capability comparison](assets/jevvision-capability-comparison.png)
+![JEVision capability overview showing its text route, visual sidecar, and typed response interface](assets/jevvision-capability-comparison.png)
 
-This earlier comparison graphic shows the input routes and context limits alongside text-only development results. Its public-panel score uses examples seen during training.
+The capability graphic summarizes the separate input routes and shared response format. Its 80,000-token value is a configured request limit, not a measure of answer quality.
 
 The serving route processed 76,999 text tokens and 76,998 image-plus-text tokens in recorded acceptance checks. The text check used the included KEV-0.8B option; the image check used the visual sidecar. These checks establish request handling near 77K, while long-context answer quality remains to be evaluated.
 
@@ -142,11 +142,11 @@ The default text adapter was fitted to public JevBench decision examples, starti
 
 ## Evaluation scope
 
-JEVision currently has functional checks for typed responses, image routing, and long request acceptance, plus the recorded real-photo example above. Independent, broad image accuracy and long-context answer quality have not yet been measured. The 80,000-token value is a service limit, and requests beyond it are rejected instead of silently truncated. No comparative accuracy, latency, or cost claim is made here.
+JEVision currently has functional checks for typed responses, image routing, and long request acceptance, plus the recorded real-photo example above. The public JevBench examples were used to fit and select the text adapter, so accuracy on those examples is not an independent benchmark result. A separate 33-question grouped text test recorded 21 correct (63.6%) with a 4,096-token cap. This small test does not measure the visual route or long-context answer quality. The 80,000-token value is a service limit, and requests beyond it are rejected instead of silently truncated. No comparative accuracy, latency, or cost claim is made here.
 
-![Text-only public and grouped test results from the earlier release](assets/jevbench-m3-results.png)
+![JEVision text evaluation scope and the recorded 21 of 33 grouped-test result](assets/jevbench-m3-results.png)
 
-This preserved results graphic separates the in-sample public panel from a small grouped holdout. It does not evaluate the visual route.
+This results graphic records the small grouped text test and distinguishes it from the public examples used during training and selection. It does not evaluate the visual route.
 
 The [manifest](jevvision_manifest.json) identifies the components, their source revisions, hashes, and serving limit. Users should assess the model on their own images and decision tasks before relying on its outputs.