Conan: Progressive Learning to Reason Like a Detective over Multi-Scale Visual Evidence
This addresses the problem of accurate visual reasoning in videos for AI systems, representing a strong domain-specific advance rather than a foundational breakthrough.
The paper tackles the challenge of multi-step video reasoning in multimodal large language models, where existing methods suffer from ungrounded conclusions or inaccurate evidence localization. The proposed Conan framework achieves state-of-the-art performance, surpassing the baseline Qwen2.5-VL-7B-Instruct by an average of over 10% in accuracy across six benchmarks.
Video reasoning, which requires multi-step deduction across frames, remains a major challenge for multimodal large language models (MLLMs). While reinforcement learning (RL)-based methods enhance reasoning capabilities, they often rely on text-only chains that yield ungrounded or hallucinated conclusions. Conversely, frame-retrieval approaches introduce visual grounding, yet still struggle with inaccurate evidence localization. To address these limitations, we present Conan, a framework for evidence-grounded multi-step video reasoning. Conan identifies context and evidence frames, reasons over cross-frame clues, and adaptively decides when to conclude or explore further. To achieve this, we 1) construct Conan-91K, a large-scale dataset of automatically generated reasoning traces that include frame identification, evidence reasoning, and action decision, and 2) design a multi-stage progressive cold-start strategy combined with an Identification-Reasoning-Action (AIR) RLVR training framework to progressively incentivize multi-step visual reasoning. Extensive experiments on six multi-step reasoning benchmarks demonstrate that Conan surpasses the baseline Qwen2.5-VL-7B-Instruct by an average of over 10% in accuracy, achieving state-of-the-art performance. Furthermore, Conan generalizes effectively to long video understanding tasks, validating its strong scalability and robustness.