org card: per-experiment hyperparameters and VQA/caption results

README.md

@@ -50,6 +50,7 @@ Winners are merged into the bf16 base and re-validated: adapters are evaluated o
 | Aspaklarya-018 | 16 / 16 | 0.0 | 2e-4 | linear | trained | 1c + 2q | 18,864 | **bf16 base** (no 4-bit) |
 | Aspaklarya-019 | 16 / 16 | 0.05 | 2e-4 | linear | trained | 1c + 2q | 18,864 | **bf16 base** + 011 recipe |
 | Aspaklarya-020 | 16 / 16 | 0.0 | 2e-4 | linear | trained | 1c + 2q | **35,981** | bf16 base + **+91% data** |
+| Aspaklarya-021 | 16 / 16 | 0.0 | 2e-4 | linear | trained | 1c + 2q | **35,981** | as 020, **8 epochs** not 4 |
 
 Constant everywhere: effective batch 32 (except 014 at 16, 015-017 at 64), warmup 0.03, grad clip 0.3 (1.0 for 011/012), seed 3407.
 
@@ -66,6 +67,7 @@ Best checkpoint per run on a fixed 1000-image draw (2000 graded items), with boo
 | Aspaklarya-018 | 82.28 ±0.69 | 82.70 ±1.51 | 82.49 ±0.86 | 80.46 ±0.49 | 75.80 ±1.11 | 78.13 ±0.63 | 78.70 ±0.67 | 69.15 ±1.61 | 73.92 ±0.90 |
 | Aspaklarya-015 | 82.14 ±0.64 | 82.76 ±1.46 | 82.45 ±0.82 | 80.07 ±0.49 | 74.73 ±1.13 | 77.40 ±0.63 | 78.07 ±0.72 | 66.98 ±1.65 | 72.52 ±0.92 |
 | Aspaklarya-006 | 81.40 ±0.71 | 82.95 ±1.50 | 82.18 ±0.90 | 79.81 ±0.50 | 75.79 ±1.12 | 77.80 ±0.65 | 78.26 ±0.68 | 68.88 ±1.63 | 73.57 ±0.91 |
+| Aspaklarya-021 | 82.14 ±0.70 | 81.93 ±1.53 | 82.03 ±0.88 | 80.93 ±0.48 | 75.41 ±1.13 | 78.17 ±0.65 | 79.76 ±0.64 | 69.12 ±1.65 | 74.44 ±0.92 |
 | Aspaklarya-005 | 81.37 ±0.67 | 82.37 ±1.50 | 81.87 ±0.85 | 79.55 ±0.49 | 75.74 ±1.11 | 77.64 ±0.62 | 77.79 ±0.71 | 69.35 ±1.65 | 73.57 ±0.93 |
 | Aspaklarya-012 | 81.68 ±0.66 | 82.05 ±1.53 | 81.86 ±0.86 | 79.73 ±0.48 | 75.02 ±1.13 | 77.37 ±0.64 | 77.85 ±0.71 | 68.23 ±1.63 | 73.04 ±0.90 |
 | Aspaklarya-014 | 81.51 ±0.67 | 81.93 ±1.48 | 81.72 ±0.83 | 79.95 ±0.49 | 74.77 ±1.13 | 77.36 ±0.63 | 78.45 ±0.71 | 67.87 ±1.66 | 73.16 ±0.92 |
@@ -160,14 +162,32 @@ Every run up to 017 trained on 6,288 images / 18,864 samples. The source table a
 Two further points matter for anyone repeating this:
 
 * **The in-domain penalty is transient.** At matched optimizer steps the extra data looked actively harmful early — step 2,360 was 1.65 behind on in-domain — and that deficit had vanished by step 4,500. Stopping early would have refuted the best run in the programme.
-* **The run never converged.** Overall score rose at all seven checkpoints, including the last. Four epochs is too short for this data volume, so the ceiling here is unknown.
+* **The climb to the final checkpoint was schedule, not headroom.** Overall score rose at all seven checkpoints including the last, which suggested an unexhausted budget. **Aspaklarya-021 tested that directly and refuted it** — see below.
 
 More data is therefore the live axis, and the only one still producing movement.
 
+## Schedule length
+
+**Aspaklarya-021** repeats 020 on the same enlarged corpus with **8 epochs instead of 4** (9,000 steps, checkpoints step-aligned with 020). It peaked at step 4,720 and then declined; the stop rule ended it at step 5,900 with 3,100 steps unspent.
+
+| step | overall | caption | vqa | in-domain |
+|---|---|---|---|---|
+| 1,180 | 77.66 | 79.19 | 76.12 | 81.56 |
+| 2,360 | 78.23 | 80.55 | 75.92 | 81.25 |
+| 3,540 | 78.37 | 80.33 | 76.41 | 82.18 |
+| **4,720** | **78.69** | 80.86 | **76.52** | 82.39 |
+| 5,900 | 77.60 | 80.75 | 74.45 | 81.15 |
+
+Its best checkpoint, paired against 020 on the same 1000 images, is negative on every slice and significant on none: overall **-0.53 ±0.43**, in-domain **-0.39 ±0.60**, RSICD **-0.66 ±0.64**.
+
+**Data volume helped; data duration did not.** Four epochs is already the right length at this corpus size.
+
+The failure mode is asymmetric and worth recording: at step 5,900 **captioning was intact (80.75) while VQA collapsed (76.52 → 74.45)**. Overtraining this recipe degrades VQA first, so a stop rule watching captioning alone would not catch it.
+
 ## Status
 
 The **hyperparameter** search is closed. Rank, alpha, dropout, learning rate, schedule, weight decay, gradient clipping, vision-tower freezing, QA-per-image, caption prompt pool, batch geometry, effective batch size, learning rate at large batch and base-weight precision were all swept across 19 recipes under identical conditions. None displaced the reference recipe by more than the paired error bar.
 
 Batch size and learning rate illustrate that ceiling. Effective batch 16 came in at **-0.59** overall, batch 64 at 2e-4 at **-0.56**, and batch 64 at 3e-4 at **-0.34** overall and **-0.87** in-domain; batch 64 at 4e-4 was stopped at epoch 4 while declining. In each case captioning held steady or improved while VQA fell.
 
-Enlarging the training set did what none of those could. The next steps follow from it: train the same corpus for longer, since run 020 had not converged, and add in-domain imagery, since the gain so far is concentrated in the general slice.
+Enlarging the training set did what none of those could, and training it longer did not extend the gain. That narrows the remaining question to **data composition**: the corpus added was RSICD-only, and the gain it produced is correspondingly RSICD-only while in-domain stays flat. Additional in-domain imagery is the targeted next step.

index.html

@@ -86,6 +86,7 @@ a { color: #0969da; }
 <tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-018</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">16 / 16</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">0.0</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">2e-4</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">linear</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">trained</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">1c + 2q</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">18,864</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap"><strong>bf16 base</strong> (no 4-bit)</td></tr>
 <tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-019</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">16 / 16</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">0.05</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">2e-4</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">linear</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">trained</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">1c + 2q</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">18,864</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap"><strong>bf16 base</strong> + 011 recipe</td></tr>
 <tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-020</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">16 / 16</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">0.0</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">2e-4</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">linear</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">trained</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">1c + 2q</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap"><strong>35,981</strong></td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">bf16 base + <strong>+91% data</strong></td></tr>
+<tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-021</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">16 / 16</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">0.0</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">2e-4</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">linear</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">trained</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">1c + 2q</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap"><strong>35,981</strong></td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">as 020, <strong>8 epochs</strong> not 4</td></tr>
 </tbody></table>
 <p>Constant everywhere: effective batch 32 (except 014 at 16, 015-017 at 64), warmup 0.03, grad clip 0.3 (1.0 for 011/012), seed 3407.</p>
 <h2>Results</h2>
@@ -109,6 +110,7 @@ a { color: #0969da; }
 <tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-018</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.28 ±0.69</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.70 ±1.51</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.49 ±0.86</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">80.46 ±0.49</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">75.80 ±1.11</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">78.13 ±0.63</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">78.70 ±0.67</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">69.15 ±1.61</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">73.92 ±0.90</td></tr>
 <tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-015</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.14 ±0.64</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.76 ±1.46</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.45 ±0.82</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">80.07 ±0.49</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">74.73 ±1.13</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">77.40 ±0.63</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">78.07 ±0.72</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">66.98 ±1.65</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">72.52 ±0.92</td></tr>
 <tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-006</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.40 ±0.71</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.95 ±1.50</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.18 ±0.90</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">79.81 ±0.50</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">75.79 ±1.12</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">77.80 ±0.65</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">78.26 ±0.68</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">68.88 ±1.63</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">73.57 ±0.91</td></tr>
+<tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-021</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.14 ±0.70</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.93 ±1.53</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.03 ±0.88</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">80.93 ±0.48</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">75.41 ±1.13</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">78.17 ±0.65</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">79.76 ±0.64</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">69.12 ±1.65</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">74.44 ±0.92</td></tr>
 <tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-005</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.37 ±0.67</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.37 ±1.50</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.87 ±0.85</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">79.55 ±0.49</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">75.74 ±1.11</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">77.64 ±0.62</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">77.79 ±0.71</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">69.35 ±1.65</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">73.57 ±0.93</td></tr>
 <tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-012</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.68 ±0.66</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.05 ±1.53</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.86 ±0.86</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">79.73 ±0.48</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">75.02 ±1.13</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">77.37 ±0.64</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">77.85 ±0.71</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">68.23 ±1.63</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">73.04 ±0.90</td></tr>
 <tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">Aspaklarya-014</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.51 ±0.67</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.93 ±1.48</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.72 ±0.83</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">79.95 ±0.49</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">74.77 ±1.13</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">77.36 ±0.63</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">78.45 ±0.71</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">67.87 ±1.66</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">73.16 ±0.92</td></tr>
@@ -203,13 +205,31 @@ a { color: #0969da; }
 <p>Two further points matter for anyone repeating this:</p>
 <ul>
 <li><strong>The in-domain penalty is transient.</strong> At matched optimizer steps the extra data looked actively harmful early — step 2,360 was 1.65 behind on in-domain — and that deficit had vanished by step 4,500. Stopping early would have refuted the best run in the programme.</li>
-<li><strong>The run never converged.</strong> Overall score rose at all seven checkpoints, including the last. Four epochs is too short for this data volume, so the ceiling here is unknown.</li>
+<li><strong>The climb to the final checkpoint was schedule, not headroom.</strong> Overall score rose at all seven checkpoints including the last, which suggested an unexhausted budget. <strong>Aspaklarya-021 tested that directly and refuted it</strong> — see below.</li>
 </ul>
 <p>More data is therefore the live axis, and the only one still producing movement.</p>
+<h2>Schedule length</h2>
+<p><strong>Aspaklarya-021</strong> repeats 020 on the same enlarged corpus with <strong>8 epochs instead of 4</strong> (9,000 steps, checkpoints step-aligned with 020). It peaked at step 4,720 and then declined; the stop rule ended it at step 5,900 with 3,100 steps unspent.</p>
+<table style="border-collapse:collapse;margin:1rem 0;font-size:.85rem;border:1px solid #d0d7de"><thead><tr>
+<th style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;background:#f6f8fa;font-weight:600;white-space:nowrap">step</th>
+<th style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;background:#f6f8fa;font-weight:600;white-space:nowrap">overall</th>
+<th style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;background:#f6f8fa;font-weight:600;white-space:nowrap">caption</th>
+<th style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;background:#f6f8fa;font-weight:600;white-space:nowrap">vqa</th>
+<th style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;background:#f6f8fa;font-weight:600;white-space:nowrap">in-domain</th>
+</tr></thead><tbody>
+<tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">1,180</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">77.66</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">79.19</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">76.12</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.56</td></tr>
+<tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">2,360</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">78.23</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">80.55</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">75.92</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.25</td></tr>
+<tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">3,540</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">78.37</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">80.33</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">76.41</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.18</td></tr>
+<tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap"><strong>4,720</strong></td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap"><strong>78.69</strong></td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">80.86</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap"><strong>76.52</strong></td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">82.39</td></tr>
+<tr><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">5,900</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">77.60</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">80.75</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">74.45</td><td style="border:1px solid #d0d7de;padding:5px 10px;text-align:left;white-space:nowrap">81.15</td></tr>
+</tbody></table>
+<p>Its best checkpoint, paired against 020 on the same 1000 images, is negative on every slice and significant on none: overall <strong>-0.53 ±0.43</strong>, in-domain <strong>-0.39 ±0.60</strong>, RSICD <strong>-0.66 ±0.64</strong>.</p>
+<p><strong>Data volume helped; data duration did not.</strong> Four epochs is already the right length at this corpus size.</p>
+<p>The failure mode is asymmetric and worth recording: at step 5,900 <strong>captioning was intact (80.75) while VQA collapsed (76.52 → 74.45)</strong>. Overtraining this recipe degrades VQA first, so a stop rule watching captioning alone would not catch it.</p>
 <h2>Status</h2>
 <p>The <strong>hyperparameter</strong> search is closed. Rank, alpha, dropout, learning rate, schedule, weight decay, gradient clipping, vision-tower freezing, QA-per-image, caption prompt pool, batch geometry, effective batch size, learning rate at large batch and base-weight precision were all swept across 19 recipes under identical conditions. None displaced the reference recipe by more than the paired error bar.</p>
 <p>Batch size and learning rate illustrate that ceiling. Effective batch 16 came in at <strong>-0.59</strong> overall, batch 64 at 2e-4 at <strong>-0.56</strong>, and batch 64 at 3e-4 at <strong>-0.34</strong> overall and <strong>-0.87</strong> in-domain; batch 64 at 4e-4 was stopped at epoch 4 while declining. In each case captioning held steady or improved while VQA fell.</p>
-<p>Enlarging the training set did what none of those could. The next steps follow from it: train the same corpus for longer, since run 020 had not converged, and add in-domain imagery, since the gain so far is concentrated in the general slice.</p>
+<p>Enlarging the training set did what none of those could, and training it longer did not extend the gain. That narrows the remaining question to <strong>data composition</strong>: the corpus added was RSICD-only, and the gain it produced is correspondingly RSICD-only while in-domain stays flat. Additional in-domain imagery is the targeted next step.</p>
 </main>
 </body>
 </html>