Imagen de Portada

SeePhys Pro: Diagnosing Modality Transfer and Blind-Training Effects in Multimodal RLVR for Physics Reasoning

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Xiang, Kun, Zhang, Terry Jingchen, Liu, Zirong, Zhou, Bokai, Tang, Yueling, Yu, Junjie, Lu, Jiacong, Huang, Shangrui, Li, Heng, Zhang, Likui, Liu, Kunkun, Zhang, Changzheng, Fang, Yangle, Guo, Boqiang, Zhen, Hui-Ling, Tu, Dandan, Huang, Yinya, Liang, Xiaodan
Formato:	Preprint
Publicado:	2026
Materias:	Artificial Intelligence
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

SeePhys: Does Seeing Help Thinking? -- Benchmarking Vision-Based Physics Reasoning
por: Xiang, Kun, et al.
Publicado: (2025)

Multimodal Reasoning for Science: Technical Report and 1st Place Solution to the ICML 2025 SeePhys Challenge
por: Liang, Hao, et al.
Publicado: (2025)

Aligning Perception, Reasoning, Modeling and Interaction: A Survey on Physical AI
por: Xiang, Kun, et al.
Publicado: (2025)

Depth-Breadth Synergy in RLVR: Unlocking LLM Reasoning Gains with Adaptive Exploration
por: Yang, Zhicheng, et al.
Publicado: (2025)

Lean Meets Theoretical Computer Science: Scalable Synthesis of Theorem Proving Challenges in Formal-Informal Pairs
por: Zhang, Terry Jingchen, et al.
Publicado: (2025)

ProPhy: Progressive Physical Alignment for Dynamic World Simulation
por: Wang, Zijun, et al.
Publicado: (2025)

AtomThink: Multimodal Slow Thinking with Atomic Step Reasoning
por: Xiang, Kun, et al.
Publicado: (2024)

Climate risk and green total factor productivity in agriculture: The moderating role of climate policy uncertainty
por: Miao Wang, et al.
Publicado: (2024)

Educators' Perceptions of Large Language Models as Tutors: Comparing Human and AI Tutors in a Blind Text-only Setting
por: Chowdhury, Sankalan Pal, et al.
Publicado: (2025)

ATG: Benchmarking Automated Theorem Generation for Generative Language Models
por: Lin, Xiaohan, et al.
Publicado: (2024)

TreeRPO: Tree Relative Policy Optimization
por: Yang, Zhicheng, et al.
Publicado: (2025)

Recycling Failures: Salvaging Exploration in RLVR via Fine-Grained Off-Policy Guidance
por: Ren, Yanwei, et al.
Publicado: (2026)

Robust H2 and H∞ Filter Design for Polytopic Systems by Dilating Matrices
por: Yuefeng Cui, et al.
Publicado: (2025)

Test of Time: Rethinking Temporal Signal of Benchmark Contamination
por: Zhang, Terry Jingchen, et al.
Publicado: (2025)

Stargazer: A Scalable Model-Fitting Benchmark Environment for AI Agents under Astrophysical Constraints
por: Liu, Xinge, et al.
Publicado: (2026)

CLOMO: Counterfactual Logical Modification with Large Language Models
por: Huang, Yinya, et al.
Publicado: (2023)

Exploring Multi-Temperature Strategies for Token- and Rollout-Level Control in RLVR
por: Zhuang, Haomin, et al.
Publicado: (2025)

Endo-G$^{2}$T: Geometry-Guided & Temporally Aware Time-Embedded 4DGS For Endoscopic Scenes
por: Liu, Yangle, et al.
Publicado: (2025)

Quantile Advantage Estimation: Stabilizing RLVR for LLM Reasoning
por: Wu, Junkang, et al.
Publicado: (2025)

EAST: Environment-Aware Stylized Transition Along the Reality-Virtuality Continuum
por: Zhang, Xiaohan, et al.
Publicado: (2025)

Multifractal analysis of the convergence exponents for the digits in $d$-decaying Gauss like dynamical systems
por: Song, Kunkun, et al.
Publicado: (2024)

PhysGame: Uncovering Physical Commonsense Violations in Gameplay Videos
por: Cao, Meng, et al.
Publicado: (2024)

FVEL: Interactive Formal Verification Environment with Large Language Models via Theorem Proving
por: Lin, Xiaohan, et al.
Publicado: (2024)

Reducing Differential Item Functioning via Process Data
por: Chen, Ling, et al.
Publicado: (2025)

An event‐triggered zonotopic Gaussian state estimator for discrete‐time stochastic multiplicative systems with zonotopic set‐membership and stochastic uncertainties
por: Ye Chen, et al.
Publicado: (2024)

ANCoEF: Asynchronous Neuromorphic Algorithm/Hardware Co-Exploration Framework with a Fully Asynchronous Simulator
por: Zhang, Jian, et al.
Publicado: (2024)

ArcPro: Architectural Programs for Structured 3D Abstraction of Sparse Points
por: Huang, Qirui, et al.
Publicado: (2025)

ORMind: A Cognitive-Inspired End-to-End Reasoning Framework for Operations Research
por: Wang, Zhiyuan, et al.
Publicado: (2025)

AlignedCoT: Prompting Large Language Models via Native-Speaking Demonstrations
por: Yang, Zhicheng, et al.
Publicado: (2023)

Critique to Verify: Accurate and Honest Test-Time Scaling with RL-Trained Verifiers
por: Yang, Zhicheng, et al.
Publicado: (2025)

SEED: A Structural Encoder for Embedding-Driven Decoding in Time Series Prediction with LLMs
por: Li, Fengze, et al.
Publicado: (2025)

Orchestrating Tokens and Sequences: Dynamic Hybrid Policy Optimization for RLVR
por: Min, Zijun, et al.
Publicado: (2026)

Does LLM Alignment Really Need Diversity? An Empirical Study of Adapting RLVR Methods for Moral Reasoning
por: Zhang, Zhaowei, et al.
Publicado: (2026)

Open-Medical-R1: How to Choose Data for RLVR Training at Medicine Domain
por: Qiu, Zhongxi, et al.
Publicado: (2025)

The Suture‐In‐Needle, Closed‐Loop Technique for Repositioning a Dislocated Akreos Adapt Intraocular Lens
por: Jingjing Zhang, et al.
Publicado: (2025)

MUSTARD: Mastering Uniform Synthesis of Theorem and Proof Data
por: Huang, Yinya, et al.
Publicado: (2024)

The Impact of Economic Growth Targets on Environmental Pollution: A Study From Chinese Cities
por: Yicheng Zhou, et al.
Publicado: (2024)

GT-HarmBench: Benchmarking AI Safety Risks Through the Lens of Game Theory
por: Cobben, Pepijn, et al.
Publicado: (2026)

MEL: Efficient Multi-Task Evolutionary Learning for High-Dimensional Feature Selection
por: Wang, Xubin, et al.
Publicado: (2024)

LvHcS52, a Litopenaeus vannamei hemocyanin-derived peptide, restricts WSSV infection by promoting phagocytosis and activating the STAT signaling pathway.
por: Zhan, Shixiong, et al.
Publicado: (2025)