Update examples

src/about.py

@@ -79,20 +79,20 @@ These benchmarks assess fundamental reasoning and knowledge capabilities of mode
 
 <div class="benchmark-table-container">
 
-| Benchmark           | Description                                                                      | Key Skills                                   |
+| Benchmark           | Description                                                                      | Domain                                  |
 |--------------------|----------------------------------------------------------------------------------|-----------------------------------------------|
-| **ARC-Easy** / **ARC-Challenge** | Multiple-choice science questions measuring scientific & commonsense reasoning. | Scientific Reasoning, Commonsense Reasoning |
-| **DROP**             | Reading comprehension benchmark emphasizing discrete reasoning steps.             | Reading Comprehension, Discrete Reasoning    |
-| **WinoGrande**        | Commonsense reasoning challenge focused on co-reference resolution.             | Commonsense Reasoning, Co-reference Resolution|
-| **GSM8K**             | Grade-school math word problems testing arithmetic & multi-step reasoning.         | Arithmetic Reasoning, Multi-step Reasoning    |
-| **HellaSwag**        | Commonsense inference task centered on action completion.                        | Commonsense Inference, Action Completion       |
-| **HumanEval**         | Evaluates code generation and reasoning in a programming context.                | Code Generation, Code Reasoning               |
-| **IFEval**            | Specialized benchmark for incremental formal reasoning.                          | Incremental Formal Reasoning                  |
-| **MATH**              | High school-level math questions requiring detailed solutions.                  | Mathematical Reasoning, Problem Solving       |
-| **MMLU** / **MMLU-Pro**| Multi-subject multiple-choice tests of advanced knowledge.                     | Broad Knowledge, Multi-domain Understanding   |
-| **GPQA-Diamond**      | Question-answering benchmark assessing deeper reasoning & knowledge linking.      | Deep Reasoning, Knowledge Linking             |
-| **MMMU** (Multi-Choice / Open-Ended) | Multilingual & multi-domain tasks testing structured & open responses.   | Multilingual Understanding, Open-form Response |
-
+| **ARC-Easy** / **ARC-Challenge** | Multiple-choice science questions measuring scientific & commonsense reasoning. | Example  |
+| **DROP**             | Reading comprehension benchmark emphasizing discrete reasoning steps.             | Example |
+| **WinoGrande**        | Commonsense reasoning challenge focused on co-reference resolution.             | Example |
+| **GSM8K**             | Grade-school math word problems testing arithmetic & multi-step reasoning.         | Example |
+| **HellaSwag**        | Commonsense inference task centered on action completion.                        |  Example |
+| **HumanEval**         | Evaluates code generation and reasoning in a programming context.                |  Example |
+| **IFEval**            | Specialized benchmark for incremental formal reasoning.                          |  Example |
+| **IFEval**            | Specialized benchmark for incremental formal reasoning.                          |  Example |
+| **MATH**              | High school-level math questions requiring detailed solutions.                  |  Example |
+| **MMLU** / **MMLU-Pro**| Multi-subject multiple-choice tests of advanced knowledge.                     |  Example |
+| **GPQA-Diamond**      | Question-answering benchmark assessing deeper reasoning & knowledge linking.      |  Example |
+| **MMMU** (Multi-Choice / Open-Ended) | Multilingual & multi-domain tasks testing structured & open responses.   |  Example |
 </div>
 
 ### 🚀 Agentic Benchmarks
@@ -103,11 +103,11 @@ These benchmarks go beyond basic reasoning and evaluate more advanced, autonomou
 
 | Benchmark              | Description                                                                 | Key Skills                                      |
 |-----------------------|-----------------------------------------------------------------------------|-------------------------------------------------|
-| **GAIA**                | Evaluates autonomous reasoning, planning, problem-solving, & multi-turn interactions. | Autonomous Reasoning, Planning, Problem Solving  |
-| [**InterCode-CTF**](https://ukgovernmentbeis.github.io/inspect_evals/evals/cybersecurity/in_house_ctf/)   | Capture-the-flag challenge focused on code interpretation & debugging.       | Code Interpretation, Generative Debugging       |
-| **GDM-In-House-CTF**    | Capture-the-flag challenge testing web application security skills.         | Web Security Skills, Security Awareness         |
-| **AgentHarm** / **AgentHarm-Benign** | Measures harmfulness of LLM agents (and benign behavior baseline).   | Harmfulness Detection, Agent Safety             |
-| **SWE-Bench**           | Tests AI agent ability to solve software engineering tasks.                 | Software Engineering Tasks, Bug Fixing          |
+| **GAIA**                | Evaluates autonomous reasoning, planning, problem-solving, & multi-turn interactions. |  Example |
+| [**InterCode-CTF**](https://ukgovernmentbeis.github.io/inspect_evals/evals/cybersecurity/in_house_ctf/)   | Capture-the-flag challenge focused on code interpretation & debugging.       |  Example |
+| **GDM-In-House-CTF**    | Capture-the-flag challenge testing web application security skills.         |  Example |
+| **AgentHarm** / **AgentHarm-Benign** | Measures harmfulness of LLM agents (and benign behavior baseline).   |  Example |
+| **SWE-Bench**           | Tests AI agent ability to solve software engineering tasks.                 |  Example |
 
 </div>
 """

src/display/css_html_js.py

@@ -104,6 +104,11 @@ custom_css = """
     overflow-x: auto;
 }
 
+.llm-benchmark-tab-table .table-wrap table.table a {
+    font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Roboto, "Helvetica Neue", Arial, sans-serif;
+    color: #ec4899
+}
+
 .llm-benchmark-tab-table .table-wrap table.table tr td,
 .llm-benchmark-tab-table .table-wrap table.table tr th {
     border-bottom: 1px solid var(--border-color-primary);
@@ -165,6 +170,7 @@ custom_css = """
     }
 }
 
+
 """
 
 get_window_url_params = """