ConvBench: A Multi-Turn Conversation Evaluation Benchmark with Hierarchical Capability for Large Vision-Language Models

TL;DR

ConvBench is a comprehensive multi-turn conversation benchmark for Large Vision-Language Models that evaluates perception, reasoning, and creativity, revealing performance gaps and guiding future improvements.

Contribution

It introduces a hierarchical multi-level evaluation benchmark for LVLMs, capturing complex cognitive capabilities and enabling precise error attribution.

Findings

Models lag behind humans in multi-turn visual dialogues.

Weak perception in models hampers reasoning and creativity.

Benchmark facilitates targeted improvements in LVLMs.

Abstract

This paper presents ConvBench, a novel multi-turn conversation evaluation benchmark tailored for Large Vision-Language Models (LVLMs). Unlike existing benchmarks that assess individual capabilities in single-turn dialogues, ConvBench adopts a three-level multimodal capability hierarchy, mimicking human cognitive processes by stacking up perception, reasoning, and creativity. Each level focuses on a distinct capability, mirroring the cognitive progression from basic perception to logical reasoning and ultimately to advanced creativity. ConvBench comprises 577 meticulously curated multi-turn conversations encompassing 215 tasks reflective of real-world demands. Automatic evaluations quantify response performance at each turn and overall conversation level. Leveraging the capability hierarchy, ConvBench enables precise attribution of conversation mistakes to specific levels. Experimental results reveal a performance gap between multi-modal models, including GPT4-V, and human performance in multi-turn conversations. Additionally, weak fine-grained perception in multi-modal models contributes to reasoning and creation failures. ConvBench serves as a catalyst for further research aimed at enhancing visual dialogues.