Add GDPval rubric scores

README.md

@@ -89,10 +89,10 @@ The model is DeepSeek-V4-Flash-0731 NVFP4 quantized with nvidia-modelopt **v0.46
 **Properties:** Undisclosed
 
 ## Evaluation Dataset:
-**Datasets:** GPQA Diamond, AA-LCR, τ²-Bench Telecom, SciCode, IFBench, Terminal-Bench v2.1 <br>
+**Datasets:** GPQA Diamond, AA-LCR, τ²-Bench Telecom, SciCode, IFBench, Terminal-Bench v2.1, GDPval <br>
 **Data Collection Method by dataset:** Hybrid: Automated, Manually-Collected <br>
 **Labeling Method by dataset:** Hybrid: Manually-Labeled, Automated <br>
-**Properties:** We evaluated the model on reasoning, long-context recall, agentic tool-use, coding, and instruction-following benchmarks: GPQA Diamond contains 448 graduate-level multiple-choice questions written by domain experts in biology, physics, and chemistry. AA-LCR (Artificial Analysis Long Context Recall) evaluates a model's ability to accurately retrieve and recall information from long input contexts. τ²-Bench Telecom evaluates agentic tool-use and policy-adherence capabilities in dual-control telecom customer-service scenarios where the model interacts with a simulated user and external tools to resolve account issues. SciCode evaluates scientific coding capabilities. IFBench is a benchmark for evaluating instruction-following capabilities across diverse and structured task constraints. Terminal-Bench v2.1 evaluates agentic software-engineering capability in a real terminal environment, where the model must plan and execute shell commands to complete tasks end to end. <br>
+**Properties:** We evaluated the model on reasoning, long-context recall, agentic tool-use, coding, and instruction-following benchmarks: GPQA Diamond contains 448 graduate-level multiple-choice questions written by domain experts in biology, physics, and chemistry. AA-LCR (Artificial Analysis Long Context Recall) evaluates a model's ability to accurately retrieve and recall information from long input contexts. τ²-Bench Telecom evaluates agentic tool-use and policy-adherence capabilities in dual-control telecom customer-service scenarios where the model interacts with a simulated user and external tools to resolve account issues. SciCode evaluates scientific coding capabilities. IFBench is a benchmark for evaluating instruction-following capabilities across diverse and structured task constraints. Terminal-Bench v2.1 evaluates agentic software-engineering capability in a real terminal environment, where the model must plan and execute shell commands to complete tasks end to end. GDPval is an AI evaluation benchmark designed to measure how well frontier models perform real, economically valuable knowledge work across major industries. <br>
 
 ## Inference:
 **Acceleration Engine:** **SGLang** and **vLLM** <br>
@@ -193,6 +193,8 @@ The accuracy benchmark results are presented in the table below:
    </td>
    <td><strong>Terminal-Bench v2.1</strong>
    </td>
+   <td><strong>GDPval (rubric)</strong>
+   </td>
   </tr>
   <tr>
    <td>MXFP4 (source)
@@ -209,6 +211,8 @@ The accuracy benchmark results are presented in the table below:
    </td>
    <td><strong>74.7</strong>
    </td>
+   <td><strong>93.0</strong>
+   </td>
   </tr>
   <tr>
    <td>NVFP4
@@ -225,10 +229,12 @@ The accuracy benchmark results are presented in the table below:
    </td>
    <td><strong>73.7</strong>
    </td>
+   <td><strong>93.2</strong>
+   </td>
   </tr>
 </table>
 
-> Baseline: [DeepSeek-V4-Flash-0731](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731). Measured at `temperature=1.0`, `top_p=1.0`, `max` reasoning effort, `max_new_tokens=384000`; τ²-Bench Telecom and Terminal-Bench v2.1 use `top_p=0.95`.
+> Baseline: [DeepSeek-V4-Flash-0731](https://huggingface.co/deepseek-ai/DeepSeek-V4-Flash-0731). Measured at `temperature=1.0`, `top_p=1.0`, `max` reasoning effort, `max_new_tokens=384000`; τ²-Bench Telecom, Terminal-Bench v2.1, and GDPval use `top_p=0.95`.
 
 ## Model Limitations:
 The base model was trained on data that contains toxic language and societal biases originally crawled from the internet. Therefore, the model may amplify those biases and return toxic responses especially when prompted with toxic prompts. The model may generate answers that may be inaccurate, omit key information, or include irrelevant or redundant text producing socially unacceptable or undesirable text, even if the prompt itself does not include anything explicitly offensive.