## Horizontal Bar Chart: Documentation Depth by Category ### Overview The image is a horizontal bar chart illustrating the documentation depth for various categories related to model cards. The chart compares the number of model cards that have "Detailed," "Mentioned Only," or "Absent" documentation for each category. The categories are listed on the vertical axis, and the number of model cards is on the horizontal axis, ranging from 0 to 100. ### Components/Axes * **Title:** Documentation Depth by Category * **X-Axis:** Number of Model Cards (scale from 0 to 100, with increments of 20) * **Y-Axis:** Categories (Model Architecture, Compute Requirements, Evaluation Metrics, License, Intended Use, Training Data, Limitations, Bias Fairness, Safety Evaluation, Out Of Scope, Interpretability) * **Legend:** Located at the top-left of the chart. * Blue: Detailed * Yellow/Orange: Mentioned Only * Red: Absent ### Detailed Analysis Here's a breakdown of the documentation depth for each category: * **Model Architecture:** * Detailed (Blue): ~75 * Mentioned Only (Yellow/Orange): ~10 * Absent (Red): ~15 * **Compute Requirements:** * Detailed (Blue): ~75 * Mentioned Only (Yellow/Orange): ~10 * Absent (Red): ~15 * **Evaluation Metrics:** * Detailed (Blue): ~75 * Mentioned Only (Yellow/Orange): ~10 * Absent (Red): ~15 * **License:** * Detailed (Blue): ~60 * Absent (Red): ~40 * **Intended Use:** * Detailed (Blue): ~60 * Mentioned Only (Yellow/Orange): ~5 * Absent (Red): ~35 * **Training Data:** * Detailed (Blue): ~60 * Mentioned Only (Yellow/Orange): ~5 * Absent (Red): ~35 * **Limitations:** * Detailed (Blue): ~60 * Mentioned Only (Yellow/Orange): ~5 * Absent (Red): ~35 * **Bias Fairness:** * Detailed (Blue): ~50 * Mentioned Only (Yellow/Orange): ~10 * Absent (Red): ~40 * **Safety Evaluation:** * Detailed (Blue): ~50 * Mentioned Only (Yellow/Orange): ~10 * Absent (Red): ~40 * **Out Of Scope:** * Detailed (Blue): ~45 * Mentioned Only (Yellow/Orange): ~5 * Absent (Red): ~50 * **Interpretability:** * Detailed (Blue): ~15 * Absent (Red): ~85 ### Key Observations * "Model Architecture," "Compute Requirements," and "Evaluation Metrics" have the highest proportion of "Detailed" documentation. * "Interpretability" has the lowest proportion of "Detailed" documentation and the highest proportion of "Absent" documentation. * "License," "Intended Use," "Training Data," "Limitations," "Bias Fairness," and "Safety Evaluation" have a significant portion of "Absent" documentation. * "Out of Scope" has nearly equal "Detailed" and "Absent" documentation. ### Interpretation The chart reveals significant variations in documentation depth across different categories of model cards. Categories like "Model Architecture," "Compute Requirements," and "Evaluation Metrics" are generally well-documented, with a large proportion of model cards providing detailed information. In contrast, "Interpretability" is poorly documented, with most model cards lacking detailed information. This suggests that interpretability aspects of models are often overlooked or not thoroughly documented. The relatively high proportion of "Absent" documentation for categories like "License," "Intended Use," "Training Data," and "Limitations" indicates potential areas for improvement in model card documentation practices. The data highlights the need for a more consistent and comprehensive approach to documenting all relevant aspects of machine learning models.

\n ## Horizontal Stacked Bar Chart: Documentation Depth by Category ### Overview This is a horizontal stacked bar chart visualizing the depth of documentation across various categories related to model cards. The chart displays the number of model cards for each category, broken down into three levels of detail: "Detailed," "Mentioned Only," and "Absent." ### Components/Axes * **Title:** "Documentation Depth by Category" (centered at the top) * **Y-axis (Vertical):** Categories of documentation. The categories, listed from top to bottom, are: * Model Architecture * Compute Requirements * Evaluation Metrics * License * Intended Use * Training Data * Limitations * Bias Fairness * Safety Evaluation * Out Of Scope * Interpretability * **X-axis (Horizontal):** Number of Model Cards. Scale ranges from 0 to 100, with tick marks at 0, 20, 40, 60, 80, and 100. * **Legend:** Located in the top-right corner. * "Detailed" - Orange * "Mentioned Only" - Yellow * "Absent" - Blue ### Detailed Analysis Each horizontal bar represents a category. The bar is divided into three colored segments representing the number of model cards with "Detailed," "Mentioned Only," and "Absent" documentation for that category. * **Model Architecture:** Approximately 82 cards have "Detailed" documentation, 12 have "Mentioned Only," and 6 are "Absent." * **Compute Requirements:** Approximately 72 cards have "Detailed" documentation, 16 have "Mentioned Only," and 12 are "Absent." * **Evaluation Metrics:** Approximately 76 cards have "Detailed" documentation, 16 have "Mentioned Only," and 8 are "Absent." * **License:** Approximately 84 cards have "Detailed" documentation, 8 have "Mentioned Only," and 8 are "Absent." * **Intended Use:** Approximately 68 cards have "Detailed" documentation, 20 have "Mentioned Only," and 12 are "Absent." * **Training Data:** Approximately 56 cards have "Detailed" documentation, 24 have "Mentioned Only," and 20 are "Absent." * **Limitations:** Approximately 76 cards have "Detailed" documentation, 16 have "Mentioned Only," and 8 are "Absent." * **Bias Fairness:** Approximately 52 cards have "Detailed" documentation, 28 have "Mentioned Only," and 20 are "Absent." * **Safety Evaluation:** Approximately 72 cards have "Detailed" documentation, 20 have "Mentioned Only," and 8 are "Absent." * **Out Of Scope:** Approximately 64 cards have "Detailed" documentation, 24 have "Mentioned Only," and 12 are "Absent." * **Interpretability:** Approximately 32 cards have "Detailed" documentation, 32 have "Mentioned Only," and 36 are "Absent." ### Key Observations * "Detailed" documentation is the most prevalent across all categories. * "Interpretability" has the lowest amount of "Detailed" documentation and the highest amount of "Absent" documentation. * "Model Architecture," "License," and "Evaluation Metrics" consistently show high levels of "Detailed" documentation. * "Training Data" and "Bias Fairness" have a relatively high proportion of "Mentioned Only" documentation compared to other categories. ### Interpretation The chart demonstrates a generally good level of documentation depth across most model card categories, with a strong emphasis on providing "Detailed" information. However, "Interpretability" stands out as an area needing significant improvement, as a substantial number of model cards lack detailed documentation in this crucial aspect. The consistent high levels of "Detailed" documentation for "Model Architecture," "License," and "Evaluation Metrics" suggest these areas are prioritized during model card creation. The relatively higher "Mentioned Only" values for "Training Data" and "Bias Fairness" might indicate that these aspects are often acknowledged but not fully elaborated upon. This data suggests a need for standardized documentation guidelines and potentially automated tools to ensure consistent and comprehensive documentation across all categories, particularly for "Interpretability." The chart highlights a potential risk: if interpretability is not well-documented, it could hinder understanding and responsible use of the models.

\n ## Horizontal Stacked Bar Chart: Documentation Depth by Category ### Overview This image is a horizontal stacked bar chart titled "Documentation Depth by Category." It visualizes the level of documentation provided across 11 different categories within a set of "Model Cards." Each bar represents a category and is segmented into three colors indicating the depth of documentation: "Detailed" (blue), "Mentioned Only" (yellow), and "Absent" (red). The total length of each bar represents 100 model cards. ### Components/Axes * **Title:** "Documentation Depth by Category" (centered at the top). * **Legend:** Located in the top-left corner of the chart area. It defines the three data series: * **Detailed:** Blue color. * **Mentioned Only:** Yellow color. * **Absent:** Red color. * **Y-Axis (Vertical):** Lists the 11 documentation categories. From top to bottom: 1. Model Architecture 2. Compute Requirements 3. Evaluation Metrics 4. License 5. Intended Use 6. Training Data 7. Limitations 8. Bias Fairness 9. Safety Evaluation 10. Out Of Scope 11. Interpretability * **X-Axis (Horizontal):** Labeled "Number of Model Cards." The scale runs from 0 to 100, with major tick marks at intervals of 20 (0, 20, 40, 60, 80, 100). ### Detailed Analysis The following table reconstructs the approximate data for each category. Values are estimated based on the bar segment lengths relative to the x-axis scale. The sum for each row is 100. | Category | Detailed (Blue) | Mentioned Only (Yellow) | Absent (Red) | | :--- | :--- | :--- | :--- | | **Model Architecture** | ~76 | ~8 | ~16 | | **Compute Requirements** | ~75 | ~9 | ~16 | | **Evaluation Metrics** | ~74 | ~10 | ~16 | | **License** | ~64 | ~1 | ~35 | | **Intended Use** | ~62 | ~3 | ~35 | | **Training Data** | ~56 | ~4 | ~40 | | **Limitations** | ~56 | ~5 | ~39 | | **Bias Fairness** | ~49 | ~5 | ~46 | | **Safety Evaluation** | ~46 | ~6 | ~48 | | **Out Of Scope** | ~44 | ~5 | ~51 | | **Interpretability** | ~18 | ~0 | ~82 | **Trend Verification:** * The "Detailed" (blue) segment shows a clear downward trend from the top category ("Model Architecture") to the bottom category ("Interpretability"). The line formed by the right edge of the blue segments slopes consistently downward. * The "Absent" (red) segment shows a corresponding upward trend, increasing in length as you move down the list of categories. * The "Mentioned Only" (yellow) segment is generally small across all categories, with no strong directional trend. ### Key Observations 1. **Best Documented Categories:** "Model Architecture," "Compute Requirements," and "Evaluation Metrics" have the highest proportion of detailed documentation, each with over 70% of model cards providing details. 2. **Worst Documented Category:** "Interpretability" is a significant outlier. It has the lowest "Detailed" score (~18) and the highest "Absent" score (~82), indicating it is rarely documented in detail and often completely missing. 3. **Binary Documentation Pattern:** For categories like "License" and "Intended Use," the "Mentioned Only" segment is very thin (~1-3). This suggests documentation tends to be either comprehensive ("Detailed") or entirely missing ("Absent"), with little middle ground. 4. **Ethical & Safety Gaps:** Categories related to ethical considerations and safety ("Bias Fairness," "Safety Evaluation," "Out Of Scope") all show less than 50% detailed documentation and have "Absent" portions approaching or exceeding 50%. ### Interpretation This chart provides a quantitative audit of transparency in AI model documentation (specifically "Model Cards"). The data suggests a strong bias towards documenting technical specifications (architecture, compute, metrics) over ethical, safety, and usage considerations. The stark gradient from top to bottom reveals a hierarchy of documentation priorities. Foundational technical details are consistently reported, while crucial context about a model's limitations, biases, safe application boundaries, and inner workings ("Interpretability") is frequently omitted. This creates a significant information asymmetry: users may know *what* the model is and *how* it was built, but have little insight into *how it behaves* in complex real-world scenarios or *how to use it responsibly*. The near absence of the "Mentioned Only" category for many items indicates an "all-or-nothing" approach to documentation. This lack of partial disclosure means that if a topic isn't covered in depth, it is likely not addressed at all, leaving dangerous knowledge gaps. The extreme case of "Interpretability" highlights a critical blind spot in the field, where the mechanisms of model decision-making are largely opaque in public documentation.

## Horizontal Bar Chart: Documentation Depth by Category ### Overview The chart visualizes the depth of documentation across 12 categories related to AI/ML model cards. Each category is represented by a horizontal bar segmented into three colors: blue ("Detailed"), orange ("Mentioned Only"), and red ("Absent"). The x-axis quantifies documentation coverage (0–100), while the y-axis lists categories in descending order of "Detailed" documentation. ### Components/Axes - **X-axis**: "Number of Model Cards" (0–100, linear scale) - **Y-axis**: Categories (Model Architecture, Compute Requirements, Evaluation Metrics, License, Intended Use, Training Data, Limitations, Bias Fairness, Safety Evaluation, Out Of Scope, Interpretability) - **Legend**: - Blue = Detailed - Orange = Mentioned Only - Red = Absent - **Spatial Layout**: - Legend positioned in the top-left corner - Bars aligned horizontally, descending from top (Model Architecture) to bottom (Interpretability) ### Detailed Analysis 1. **Model Architecture**: - Detailed: ~75 (blue) - Mentioned Only: ~10 (orange) - Absent: ~15 (red) 2. **Compute Requirements**: - Detailed: ~75 (blue) - Mentioned Only: ~10 (orange) - Absent: ~15 (red) 3. **Evaluation Metrics**: - Detailed: ~70 (blue) - Mentioned Only: ~10 (orange) - Absent: ~20 (red) 4. **License**: - Detailed: ~65 (blue) - Mentioned Only: ~5 (orange) - Absent: ~30 (red) 5. **Intended Use**: - Detailed: ~60 (blue) - Mentioned Only: ~5 (orange) - Absent: ~35 (red) 6. **Training Data**: - Detailed: ~55 (blue) - Mentioned Only: ~5 (orange) - Absent: ~40 (red) 7. **Limitations**: - Detailed: ~55 (blue) - Mentioned Only: ~5 (orange) - Absent: ~40 (red) 8. **Bias Fairness**: - Detailed: ~50 (blue) - Mentioned Only: ~5 (orange) - Absent: ~45 (red) 9. **Safety Evaluation**: - Detailed: ~45 (blue) - Mentioned Only: ~5 (orange) - Absent: ~50 (red) 10. **Out Of Scope**: - Detailed: ~45 (blue) - Mentioned Only: ~5 (orange) - Absent: ~50 (red) 11. **Interpretability**: - Detailed: ~20 (blue) - Mentioned Only: ~0 (orange) - Absent: ~80 (red) ### Key Observations - **Trend Verification**: - Blue bars ("Detailed") decrease monotonically from top to bottom (75 → 20). - Red bars ("Absent") increase monotonically (15 → 80). - Orange bars ("Mentioned Only") remain consistently low (~5–10) except for License (~5) and Intended Use (~5). - **Notable Outliers**: - Interpretability has the lowest "Detailed" documentation (~20) and highest "Absent" (~80). - License and Intended Use show minimal "Mentioned Only" documentation. ### Interpretation The data suggests a systematic decline in detailed documentation quality as categories progress from technical specifications (Model Architecture) to ethical considerations (Interpretability). Critical gaps appear in: 1. **Ethical/Ethical-Related Categories**: - Bias Fairness, Safety Evaluation, and Interpretability show significant "Absent" documentation (>45). 2. **Implementation-Specific Categories**: - Compute Requirements and Evaluation Metrics maintain high "Detailed" coverage but lack depth in "Mentioned Only" segments. 3. **Structural Gaps**: - License and Interpretability categories exhibit the largest documentation voids, raising concerns about compliance and model transparency. The chart underscores a prioritization of technical documentation over ethical and operational aspects, with Interpretability being the most neglected area. This pattern may reflect resource allocation biases or underestimation of interpretability's importance in model deployment.