Search papers, labs, and topics across Lattice.
FCMBench-Video, a new benchmark, is introduced to evaluate document perception, temporal grounding, and evidence-grounded reasoning in document videos using realistic capture conditions. The benchmark consists of 1,200 long-form videos composed from 495 atomic videos, paired with 11,322 expert-annotated question-answer instances in both Chinese and English. Evaluations of nine Video-MLLMs on FCMBench-Video reveal meaningful performance separation across tasks like counting, cross-document validation, and visual prompt injection, highlighting the benchmark's ability to probe evidence integration and robustness.
Video-MLLMs struggle to integrate evidence across frames in document videos, especially for tasks like cross-document validation, revealing a critical gap in their ability to reason about authenticity and prevent fraud.
Document understanding is a critical capability in financial credit review, onboarding, and remote verification, where both decision accuracy and evidence traceability matter. Compared with static document images, document videos present a temporally redundant and sequentially unfolding evidence stream, require evidence integration across frames, and preserve acquisition-process cues relevant to authenticity-sensitive and anti-fraud review. We introduce FCMBench-Video, a benchmark for document-video intelligence that evaluates document perception, temporal grounding, and evidence-grounded reasoning under realistic capture conditions. For privacy-compliant yet realistic data at scale, we organize construction as an atomic-acquisition and composition workflow that records reusable single-document clips, applies controlled degradations, and assembles long-form multi-document videos with prescribed temporal spans. FCMBench-Video is built from 495 atomic videos composed into 1,200 long-form videos paired with 11,322 expert-annotated question--answer instances, covering 28 document types over 20s--60s duration tiers and 5,960 Chinese / 5,362 English instances. Evaluations on nine recent Video-MLLMs show that FCMBench-Video provides meaningful separation across systems and capabilities: counting is the most duration-sensitive task, Cross-Document Validation and Evidence-Grounded Selection probe higher-level evidence integration, and Visual Prompt Injection provides a complementary robustness dimension. The overall score distribution is broad and approximately bell-shaped, indicating a benchmark that is neither saturated nor dominated by trivial cases. Together, these results position FCMBench-Video as a reproducible benchmark for tracking Video-MLLM progress on document-video understanding and probing capability boundaries in authenticity-sensitive credit-domain applications.
