Merge branch 'develop' into 'fb-dia-1716'

Workflow run: https://github.com/HumanSignal/label-studio/actions/runs/13021748106

docs/source/guide/ask_ai.md

@@ -0,0 +1,49 @@
+---
+title: About Ask AI - Beta 🧪
+short: Ask AI
+tier: enterprise
+type: guide
+order: 0
+order_enterprise: 356
+meta_title: About Ask AI 
+meta_description: Information about using the Ask AI tool in Label Studio
+section: "Manage Your Organization"
+date: 2025-01-28 16:40:16
+---
+
+The Ask AI tool is an LLM that has been trained on our documentation, codebase, and several other Label Studio resources. 
+
+You can use Ask AI to create or refine your labeling configuration. Instead of manually building labeling interfaces or project instructions from scratch, you can prompt the AI with a description of what your labeling project needs, and the AI generates a suggested configuration. 
+
+
+## What models do you use?
+
+We use a combination of Open Source LLMs hosted by HumanSignal and hosted LLMs from OpenAI's GPT series of models.
+
+## What data is used to generate answers?
+
+Only the data you provide in the chat window is used to generate responses, this includes:
+
+• All chat messages
+• Task samples you have uploaded
+• The current labeling config
+
+Only the chat history, tasks, and labelling configuration from the current project are used.
+
+## How is my data used to train models?
+
+The model is trained on public resources like our documentation, codebase, blog, and website. We also train it on additional examples we've written specifically for the the model. 
+
+We are on a Tier 5 OpenAI account, and opt out of all requests for training data.
+
+We track requests to Ask AI to use as quality control, but they are only used to test the outputs of the model.  For example: If Company X asks "Make a project to label opossums," we may notice that Ask AI did not provide a satisfactory answer and we may create configurations with opossums in them and train the model on them. The model would not be trained on any of Company X's questions or data.
+
+All tracked data is covered by our industry-leading [security and privacy policies](https://humansignal.com/security/).
+
+## Can I disable Ask AI?
+
+Yes, contact us to disable this feature.
+
+## HIPAA compliance
+
+Your Business Associate Agreement (BAA) does not cover these features. If you are required to comply with HIPAA, we recommend that you disable Ask AI.

docs/source/guide/release_notes/onprem/2.20.0.md

@@ -0,0 +1,63 @@
+---
+hide_sidebar: true
+---
+
+## Label Studio Enterprise 2.20.0
+
+<div class="onprem-highlight">Taxonomy for labeling, new audio hotkey, performance improvements, bug fixes </div>
+
+*Jan 28, 2025*
+
+Helm Chart version: 1.9.0
+
+### Enhancements
+
+#### Use Taxonomy for labeling
+
+There is a new `labeling` parameter available for the Taxonomy tag. When set to true, you can apply your Taxonomy classes to different regions in text. For more information, see [Taxonomy as a labeling tool](/templates/taxonomy).
+
+![Screenshot of taxonomy as labeling tool](/images/releases/2-20-taxonomy.png)
+
+#### Hotkeys for audio labeling
+
+There is a new hotkey available for pausing and starting audio: `ctrl`+`p` (Windows) or `command`+`p` (Mac). This is in addition to the space hotkey that performs the same function. However, this new hotkey is useful when you are working with audio and have a text area field in focus.
+
+#### Video frame classification template
+
+The video frame classification template is now available in the Label Studio app as well as [the documentation](/templates/video_frame_classification).
+
+
+#### Performance improvements
+
+Optimized the API calls made from the frontend within the members management and Data Manager users lists. Also optimized the Projects page for faster load times. 
+
+### Security
+
+- Upgraded pyarrow to address vulnerabilities in older packages.
+
+- Updated the default settings for CSRF cookie to be more secure and added an environment setting to control cookie age.
+
+
+### Bug fixes
+
+- Fixed an issue where the Label Studio version as displayed in the side menu was not formatted properly.
+
+- Fixed an issue where the `contextlog` was not reporting the `content_type`.
+
+- Fixed an issue with overlapping relations on the overlay on highlighting.
+
+- Fixed an issue where task IDs were being duplicated when importing a large number of tasks through the API.
+
+- Fixed an issue where users were not being redirected to the appropriate page after logging in.
+
+- Fixed an issue where users were unable to edit meta information that they previously added to bounding box regions.
+
+- Fixed multiple issues resulting from Poetry/Poetry core 2 release.
+
+- Fixed an issue where the django-rq admin page was unavailable.
+
+- Fixed a possible race condition when dynamically loading the Data Manager or editor that would prevent either from loading.
+
+- Fixed an issue where skipped tasks were not being calculated as completed when the project Skip Queue setting was set to **Ignore Skipped**.
+
+

docs/source/templates/gallery_llm_evals.ejs

@@ -41,4 +41,10 @@ cards:
   - Retrieval-Augmented Generation
   image: "/images/templates/evaluate-rag-automated-metrics.png"
   url: "/templates/llm_ragas.html"
+
+- title: Multi-Turn Chat Evaluation
+  categories:
+  - Response Evaluation
+  image: "/images/templates/multi-turn-chat.png"
+  url: "/templates/multi_turn_chat.html"
 ---

docs/source/templates/multi-page-document-annotation.md

@@ -21,7 +21,7 @@ Multi-page document annotation requires that you first pre-process your document
 To improve performance, we preload images by downloading an image via XHR and storing it in the local browser’s storage. For this to function properly, you must set up a CORS policy on the CDN/storage where the images are located. Specifically, the following headers must be set:
 
 ```
-Access-Control-Allow-Origin: https://app.heartex.com
+Access-Control-Allow-Origin: https://app.humansignal.com
 Access-Control-Allow-Methods: GET, OPTIONS
 Access-Control-Allow-Headers: Content-Type
 ```