Update ocr-vllm-judge.py: improved eval prompt + --save-results flag

- Rewrite evaluation prompt to prioritize faithfulness over formatting,
heavily penalizing added commentary/interpretation
- Add --save-results flag to push judge results (comparisons, leaderboard,
metadata) to HF Hub dataset repo

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

ocr-vllm-judge.py

@@ -101,11 +101,21 @@ def image_to_base64(image: Image.Image) -> str:
 # --- Judge prompt ---
 PAIRWISE_PROMPT = """You are an expert OCR quality evaluator. You are given a document image and TWO OCR outputs (A and B) extracted from that same image.
 
-Compare them and decide which extraction is better overall. Consider:
-- Accuracy: correct characters, no hallucinations
-- Completeness: all text captured
-- Formatting: clean structure (ignore bounding box tags like <|ref|> <|det|> if present)
-- Reading order: natural flow
+Compare them and decide which extraction is better overall.
+
+Evaluation criteria (in priority order):
+
+1. Faithfulness: The output must ONLY contain text from the document. Any added commentary, interpretation, or notes (e.g. "it appears the text says...", "the document contains...") is a serious error. Penalize heavily.
+
+2. Completeness: ALL visible text must be captured — headers, footers, marginalia, stamps, handwritten notes. Missing any section of text is a significant penalty.
+
+3. Accuracy: Correct characters, no hallucinated words or garbled text.
+
+4. Reading order: Text flows naturally as a human would read the document.
+
+5. Formatting: Clean structure. Ignore bounding box tags like <|ref|> <|det|> if present. Do NOT prefer fancier markdown formatting — plain accurate text is better than nicely formatted but incomplete text.
+
+If both outputs capture the same text with similar accuracy, respond with "tie". Only pick a winner when there is a clear quality difference.
 
 Output A:
 ---
@@ -680,6 +690,11 @@ Examples:
         default=0.85,
         help="vLLM GPU memory fraction (default: 0.85)",
     )
+    parser.add_argument(
+        "--save-results",
+        default=None,
+        help="Push judge results to this HF dataset repo (e.g. davanstrien/ocr-bench-rubenstein-judge)",
+    )
     args = parser.parse_args()
 
     # --- CUDA check ---
@@ -753,6 +768,56 @@ Examples:
     )
     print_elo_leaderboard(elo, wins, losses, ties)
 
+    # --- Save results ---
+    if args.save_results:
+        import datetime
+
+        console.print(f"\n[bold]Saving results to:[/bold] {args.save_results}")
+
+        # Comparisons config
+        comp_ds = Dataset.from_list(comparison_log)
+        comp_ds.push_to_hub(args.save_results, config_name="comparisons")
+        console.print(f"  Pushed [cyan]comparisons[/cyan] ({len(comparison_log)} rows)")
+
+        # Leaderboard config
+        ranked = sorted(elo.items(), key=lambda x: x[1], reverse=True)
+        leaderboard_rows = []
+        for model, rating in ranked:
+            total = wins[model] + losses[model] + ties[model]
+            leaderboard_rows.append(
+                {
+                    "model": model,
+                    "elo": round(rating),
+                    "wins": wins[model],
+                    "losses": losses[model],
+                    "ties": ties[model],
+                    "win_pct": round(wins[model] / total * 100) if total > 0 else 0,
+                }
+            )
+        Dataset.from_list(leaderboard_rows).push_to_hub(
+            args.save_results, config_name="leaderboard"
+        )
+        console.print(
+            f"  Pushed [cyan]leaderboard[/cyan] ({len(leaderboard_rows)} rows)"
+        )
+
+        # Metadata config
+        metadata_row = {
+            "source_dataset": args.dataset,
+            "judge_models": json.dumps(judge_models),
+            "seed": args.seed,
+            "max_samples": args.max_samples or len(ds),
+            "total_comparisons": len(comparisons),
+            "valid_comparisons": len(comparison_log),
+            "timestamp": datetime.datetime.now(datetime.timezone.utc).isoformat(),
+            "from_prs": args.from_prs,
+        }
+        Dataset.from_list([metadata_row]).push_to_hub(
+            args.save_results, config_name="metadata"
+        )
+        console.print("  Pushed [cyan]metadata[/cyan]")
+        console.print(f"  [green]Results saved to: {args.save_results}[/green]")
+
     # --- Sample comparisons ---
     console.print("\n[bold]Sample comparisons:[/bold]")
     for entry in comparison_log[:5]: