Perception First: A Frontier Native-Video Model with Self-Consistency for Implicit Video Question Answering

We describe our submission to the VRR Challenge @ CVPR 2026, built on the ImplicitQA / VRR-QA benchmark~citeimplicitqa: multiple-choice video question answering in which answers are deliberately not observable in any single frame and must be inferred from spatial layout, motion, depth, viewpoint, causality, and social context across discontinuous frames of creative video. We conduct a systematic, training-free study spanning open-source Video-LMMs (Qwen2.5-VL~citeqwen25vl, Qwen3-VL~citeqwen3vl, InternVL3, Gemma-3, and the RL-tuned video reasoners Video-R1~citevideor1 and VideoChat-R1.5~citevideochatr15) and a battery of inference-time strategies (chain-of-thought, question decomposition, describe-then-reason cascades, audio transcripts, spatial state prompting, self-consistency~citeselfconsistency, multi-model ensembling, and category routing). Our central finding is that this benchmark is perception-bound rather than reasoning-bound: reasoning-side augmentations are neutral-to-harmful, whereas base-model perceptual capability and lightweight test-time denoising are the only reliable levers. A per-category error analysis localizes the difficulty to low-level perception -- relative depth, viewpoint, and counting are the hardest categories, while causal and social reasoning are nearly solved -- and a prompt that explicitly injects monocular depth cues to attack the weakest category lowers test accuracy by 5.8 points, confirming that the model needs a better percept, not a better procedure.