PhyBench: A Physical Commonsense Benchmark for Evaluating Text-to-Image Models
read the original abstract
Text-to-image (T2I) models have made substantial progress in generating images from textual prompts. However, they frequently fail to produce images consistent with physical commonsense, a vital capability for applications in world simulation and everyday tasks. Current T2I evaluation benchmarks focus on metrics such as accuracy, bias, and safety, neglecting the evaluation of models' internal knowledge, particularly physical commonsense. To address this issue, we introduce PhyBench, a comprehensive T2I evaluation dataset comprising 700 prompts across 4 primary categories: mechanics, optics, thermodynamics, and material properties, encompassing 31 distinct physical scenarios. We assess 6 prominent T2I models, including proprietary models DALLE3 and Gemini, and demonstrate that incorporating physical principles into prompts enhances the models' ability to generate physically accurate images. Our findings reveal that: (1) even advanced models frequently err in various physical scenarios, except for optics; (2) GPT-4o, with item-specific scoring instructions, effectively evaluates the models' understanding of physical commonsense, closely aligning with human assessments; and (3) current T2I models are primarily focused on text-to-image translation, lacking profound reasoning regarding physical commonsense. We advocate for increased attention to the inherent knowledge within T2I models, beyond their utility as mere image generation tools. The data will be available soon.
This paper has not been read by Pith yet.
Forward citations
Cited by 14 Pith papers
-
PhysInOne: Visual Physics Learning and Reasoning in One Suite
PhysInOne is a new dataset of 2 million videos across 153,810 dynamic 3D scenes covering 71 physical phenomena, shown to improve AI performance on physics-aware video generation, prediction, property estimation, and m...
-
Do Image Editing Models Understand Lighting?
New 3DLP benchmark with real-world 1K HDR pairs shows state-of-the-art image editing models vary in physical lighting consistency, with best models close to reality but error-prone in low-light regions.
-
Can Image Models Imagine Time? ImageTime: A Novel Benchmark for Probing Visual World Modeling Through Spatiotemporal Consistency
ImageTime is a benchmark that probes image generation models' visual world modeling by requiring coherent four-state sequences in single images, scored via VLM judge.
-
VideoASMR-Bench: Can AI-Generated ASMR Videos Fool VLMs and Humans?
VideoASMR-Bench shows state-of-the-art VLMs fail to reliably detect AI-generated ASMR videos from real ones, though humans can still identify the fakes relatively easily.
-
Reasoning to Edit: Hypothetical Instruction-Based Image Editing with Visual Reasoning
Presents Reason50K dataset and ReasonBrain framework for hypothetical instruction-based image editing that requires physical, temporal, causal, and story reasoning.
-
WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation
Text-to-image models show significant limitations in integrating world knowledge, as measured by the new WISE benchmark and WiScore metric across 20 models.
-
T2I-FactualBench: Benchmarking the Factuality of Text-to-Image Models with Knowledge-Intensive Concepts
T2I-FactualBench is a new three-tier benchmark for factuality of knowledge-intensive concepts in T2I models, using multi-round VQA evaluation to show SOTA models need improvement.
-
Qwen-Image-Agent: Bridging the Context Gap in Real-World Image Generation
Qwen-Image-Agent is a unified agent framework that progressively builds sufficient generation context for T2I models via Context-Aware Planning and Context Grounding, achieving SOTA on IA-Bench, Mindbench, and WISE-Verified.
-
Qwen-Image-Agent: Bridging the Context Gap in Real-World Image Generation
Qwen-Image-Agent bridges the context gap in text-to-image models via Context-Aware Planning and Context Grounding that integrate plan, reason, search, memory and feedback, achieving SOTA on IA-Bench and related benchmarks.
-
Qwen-Image-Bench: From Generation to Creation in Text-to-Image Evaluation
Qwen-Image-Bench introduces a hierarchical creator-centric benchmark with 1000 prompts, 23 sub-capabilities, and a Q-Judger model that scores images on 56 verifiable facets to distinguish T2I models on fidelity and cr...
-
Training-Trajectory-Aware Token Selection
Training-Trajectory-Aware Token Selection (T3S) reconstructs the token-level training objective to overcome a performance bottleneck in continual distillation of reasoning capabilities from large to small language models.
-
FireScope: Wildfire Risk Raster Prediction with a Chain-of-Thought Oracle
FireScope is a VLM framework that generates wildfire risk rasters together with reasoning traces, showing improved cross-continental generalization when trained on US expert maps and tested on European fire events.
-
Visual Generation in the New Era: An Evolution from Atomic Mapping to Agentic World Modeling
Visual generation models are evolving from passive renderers to interactive agentic world modelers, but current systems lack spatial reasoning, temporal consistency, and causal understanding, with evaluations overemph...
-
FireScope: Wildfire Risk Raster Prediction with a Chain-of-Thought Oracle
FireScope trains a VLM on US data to output wildfire risk rasters with reasoning traces and shows improved cross-continental performance on European events compared with prior approaches.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.