Triplet Incremental Training Study - OmniDrive BLEU: leaderboard

Triplet Incremental Training v1: unchanged zero-shot InternVL3.5-4B on the source native driving-language panel; exact inference prompt/toolkit/score normalization not reported. No SFT Baseline/Ours score is assigned to this checkpoint. Source singleton measurements remain evidence; no ranking eligibility is inferred. Metric: Native BLEU score Higher is better.

Metric: Native BLEU score. Source: arxiv.org. 1 model tracked. Updated 14 Nov 2026.

Top models

#ModelScoreOverall rank
1InternVL3.5-4B25.64#1916

Interactive version: theaggregate.ai/benchmark?slug=triplet-incremental-training-study-omnidrive-bleu · How It Works · Data refreshed daily, snapshot 2026-11-16.