REVIEW 17 cited by
PhyX: Does Your Model Have the "Wits" for Physical Reasoning?
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
PhyX: Does Your Model Have the "Wits" for Physical Reasoning?
read the original abstract
Existing benchmarks fail to capture a crucial aspect of intelligence: physical reasoning, the integrated ability to combine domain knowledge, symbolic reasoning, and understanding of real-world constraints. To address this gap, we introduce PhyX: the first large-scale benchmark designed to assess models capacity for physics-grounded reasoning in visual scenarios. PhyX includes 3K meticulously curated multimodal questions spanning 6 reasoning types across 25 sub-domains and 6 core physics domains: thermodynamics, electromagnetism, mechanics, modern physics, optics, and wave\&acoustics. In our comprehensive evaluation, even state-of-the-art models struggle significantly with physical reasoning. GPT-4o, Claude3.7-Sonnet, and GPT-o4-mini achieve only 32.5%, 42.2%, and 45.8% accuracy respectively-performance gaps exceeding 29% compared to human experts. Our analysis exposes critical limitations in current models: over-reliance on memorized disciplinary knowledge, excessive dependence on mathematical formulations, and surface-level visual pattern matching rather than genuine physical understanding. We provide in-depth analysis through fine-grained statistics, detailed case studies, and multiple evaluation paradigms to thoroughly examine physical reasoning capabilities. To ensure reproducibility, we implement a compatible evaluation protocol based on widely-used toolkits such as VLMEvalKit, enabling one-click evaluation. More details are available on our project page: https://phyx-bench.github.io/.
Forward citations
Cited by 17 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...
-
FeynmanBench: Benchmarking Multimodal LLMs on Diagrammatic Physics Reasoning
FeynmanBench is the first benchmark for evaluating multimodal LLMs on diagrammatic reasoning with Feynman diagrams, revealing systematic failures in enforcing physical constraints and global topology.
-
Apple-$\pi$: Benchmarking Thinking with Video Towards Law-Grounded Physical Intelligence
A law-grounded benchmark, Apple-PI, grades video models stage-by-stage on physics reasoning and finds they top out at 0.473, well short of reliable simulation.
-
SeePhys Pro: Diagnosing Modality Transfer and Blind-Training Effects in Multimodal RLVR for Physics Reasoning
Multimodal AI models for physics reasoning lose performance when information shifts from text to images, and RLVR training gains often come from non-visual textual or distributional cues rather than actual visual evidence.
-
SeePhys Pro: Diagnosing Modality Transfer and Blind-Training Effects in Multimodal RLVR for Physics Reasoning
SeePhys Pro benchmark reveals multimodal models degrade on physics reasoning as information transfers from text to images, with blind training improvements often stemming from textual cues rather than visual evidence.
-
FeynmanBench: Benchmarking Multimodal LLMs on Diagrammatic Physics Reasoning
Across 19 multimodal LLMs, local Feynman-diagram recognition stays high while topological reconstruction and full amplitude derivation collapse.
-
PolyReal: A Benchmark for Real-World Polymer Science Workflows
PolyReal benchmark shows leading MLLMs perform well on polymer knowledge reasoning but drop sharply on practical tasks like lab safety analysis and raw data extraction.
-
PerceptionBench: Evaluating Atomic Visual Perception in Multimodal Large Language Models
Frontier MLLMs remain far from mastering atomic visual perception: none reach 60% on a failure-derived, perception-only benchmark of ten capabilities.
-
Trace: A Taxonomy-Guided Environment for Multidomain Visual Reasoning
RLVR training on 64,000 procedurally generated Trace instances improves Qwen2.5-VL macro-average on 24 external visual reasoning benchmarks by 3.51 points at 3B and 4.06 points at 7B.
-
Reasoning Dynamics and the Limits of Monitoring Modality Reliance in Vision-Language Models
VLMs show answer inertia in CoT reasoning and remain influenced by misleading textual cues even with sufficient visual evidence, making CoT an incomplete window into modality reliance.
-
Cognitive Pivot Points and Visual Anchoring: Unveiling and Rectifying Hallucinations in Multimodal Reasoning Models
Hallucinations in multimodal reasoners cluster at high-entropy pivot tokens due to intermediate-layer visual anchoring failure, and V-STAR mitigates this with attention rewards plus forced reflection.
-
Cognitive Mismatch in Multimodal Large Language Models for Discrete Symbol Understanding
MLLMs exhibit a consistent recognition-reasoning inversion on discrete visual symbols across domains, underperforming on elementary perception while appearing competent on higher-level reasoning via linguistic compensation.
-
VisPhyWorld: Probing Physical Reasoning via Code-Driven Video Reconstruction
VisPhyWorld evaluates MLLMs' physical reasoning via executable code generation for video reconstruction, with VisPhyBench showing strong semantics but weak parameter inference and dynamics simulation.
-
Inferring Dynamic Physical Properties from Video Foundation Models
Video foundation models infer dynamic physical properties such as elasticity, viscosity, and friction from videos at levels close to classical oracles while outperforming current MLLMs with suitable prompting.
-
Cognitive Pivot Points and Visual Anchoring: Unveiling and Rectifying Hallucinations in Multimodal Reasoning Models
Multimodal reasoning models hallucinate at high-entropy cognitive bifurcation points due to loss of visual semantic anchoring, and the V-STAR training paradigm with HVAR rewards and FRM reflection mitigates this by re...
-
Though Language Models Err While They Strive: Conformal Prediction for Self-Correcting Scientific Generation
A graph-based conformal wrapper that filters and regenerates LLM reasoning steps claims formal coverage guarantees on scientific validity, but its evaluation is circular and its gains are confounded with sampling effo...
-
Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity
Seed2.0 model series reports gains in reasoning, visual understanding, search, and reliability on intricate long-horizon tasks via an internal evaluation system.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.