跨领域第一视角视频推理新框架，解决时序模糊与语义差异难题
OmniEgo-R$^2$: A Routed Reasoning Framework for the 1st Cross-Domain EgoCross Challenge at CVPR 2026
- 设计路由推理流程，动态适配不同场景的视觉语义规律
- 在有限源和开放源赛道分别取得66.35%与66.77%准确率，排名第二
- 适合需要跨场景理解的医疗、工业、极限运动等领域应用
CVPR 2026 EgoVis首届跨域第一视角挑战赛评估多模态大模型在手术、工业、极限运动及动物视角视频中的推理能力。我们于源受限与开放源双赛道均获第二名。本文将EgoCross视为鲁棒的跨域具身视频推理问题，而非简单选择题问答。识别出三大挑战：(C1) 时间边界模糊，关键状态转换稀疏采样且常发生于帧间；(C2) 跨域语义粒度不匹配，同一能力需不同领域视觉语法表达；(C3) 近似选项下决策不稳定，长程多模态推理易选无效干扰项或生成错误输出。为此提出OmniEgo-R$^2$（全领域第一视角路由推理），统一包含时间证据归一化、领域无关能力路由、结构化感知-动力-决策推理、边界感知选项验证与防御性答案校准的推理流水线。以Qwen3-VL-4B-SFT作为各域视觉语言主干，外挂轻量级测试时推理与解析程序。最终提交结果在源受限赛道达66.35%总体准确率，开放源赛道达66.77%，双榜排名第二。代码已开源：https://github.com/Lee-zixu/OmniEgo-R2
原文摘要 · Abstract (English)
The 1st Cross-Domain EgoCross Challenge at EgoVis, CVPR 2026 evaluates whether multimodal large language models can reason over egocentric videos across surgery, industry, extreme sports, and animal perspective. We achieved second place in both the Source-Limited and Open-Source tracks. In this report, we formulate EgoCross as a robust cross-domain embodied video reasoning problem rather than a simple multiple-choice visual question answering task. We identify three key challenges: (C1) temporal boundary ambiguity, where critical state transitions are sparsely sampled and often occur between frames; (C2) cross-domain semantic granularity mismatch, where the same capability requires different domain-specific visual grammar; and (C3) decision instability under close options, where long multimodal reasoning can select unsupported distractors or produce malformed outputs. To address them, we propose OmniEgo-R$^2$ (Omnidomain Egocentric Routed Reasoning), a unified routed reasoning pipeline consisting of temporal-evidence normalization, domain-agnostic capability routing, structured perception--dynamics--decision reasoning, boundary-aware option verification, and defensive answer calibration. OmniEgo-R$^2$ uses the Qwen3-VL-4B-SFT checkpoints on each EgoCross domain as the visual-language backbone, and wraps them with lightweight test-time reasoning and parsing programs. Our final submissions obtain 66.35% overall accuracy in the Source-Limited track and 66.77% in the Open-Source track, ranking second in both leaderboards. The codes are available on https://github.com/Lee-zixu/OmniEgo-R2
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。