Clamping CoT probabilities to 40-60%

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• Clamping CoT probabilities to 40-60% range for RL-CAI with CoT improves robustness and reduces extreme responses.finding0.830
  Section 4.3 describes clamping at 40-60 led to better behavior than clamping at 20-80.
• VS-CoT with GPT-4.1 as generator achieves 45.9% accuracy on Qwen3-1.7B-Base fine-tuned on math benchmarks, the highest single resultfinding0.710
  Best performing VS variant for math synthetic data generation with GPT-4.1
• 90.45% accuracy on ScienceQA benchmark with Multimodal-CoT Large (738M parameters)finding0.686
  State-of-the-art result on ScienceQA; represents +3.91% improvement over prior best published result of 86.54%.
• One-stage CoT (QCM→RA) shows 12.31% accuracy drop vs. no-CoT (QCM→A) on ScienceQA; two-stage framework (rationale generation + answer inference) achieves 85.31% accuracy with vision featuresfinding0.686
  Empirical evidence that naive one-stage CoT fails in language-only setting; two-stage + vision achieves state-of-the-art.
• Verbalized probabilities for programming languages show only Pearson r=0.182 (GPT-4.1) correlation with corpus frequencies, indicating weaker calibrationfinding0.681
  Limits of verbalized probability calibration when corpus frequency and perceived popularity diverge
• VS-CoT achieves the highest Top@1 accuracy (0.348) and Pass@N accuracy (0.485) on SimpleQA among all methods testedfinding0.678
  Shows VS not only maintains but can slightly improve factual accuracy compared to baseline methods
• CoT improves in-distribution but may harm out-of-distribution generalizationclaim0.675
  Interpretation of scope generalization results
• Bootstrap 95% CI for mean contemplative lift: +2.62 [2.16, 2.90]; baseline rank concordance under perturbation: 0.909; top-5 stability: 89.6%finding0.673
  Validates robustness of universal lift finding