跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-17AI 評分39

Axiomatic World Modeling for Physics Reasoning:以公理化世界建模提升物理推理

Axiomatic World Modeling for Physics Reasoning

AI 導讀

RLAWM 透過懲罰模型內部公理化世界模型與題目所指定物理世界之間的偏差,提升 LLM 的物理推理表現。方法在訓練時以建模和解題兩個層級分別評分,推理時不增加額外成本;在 PhysReason 和 PHYSICS 上分別較基線提升最多 +12.2 和 +5.2 分。 RLAWM 在奧林匹克級物理和數學題上較最強基線提升 2.45x 和 +10 分。

正文

Published on Oct 17

Authors:

,

,

Abstract

RLAWM improves physics reasoning by penalizing divergence between a model's internal world model and true physical axioms through hierarchical modeling and solving rewards.

Recent research-level physics-reasoning benchmarks for LLMs reveal a failure mode that lies not in knowledge or execution, but in modeling the world being reasoned about. Models often silently alter given quantities, fabricate laws, or inject unstated assumptions to force a problem into familiar patterns, and then reason inside this corrupted world. We define this premise-level divergence as modeling drift: the gap between a model's internal axiomatic world model (AWM) and the true physical world specified in the problem. Although RLVR and PRM have yielded substantial gains in mathematics and code, their reward shapes do not directly supervise world modeling in axiomatic domains such as physics. We propose Reinforcement Learning with Axiomatic World Modeling (RLAWM), which optimizes the policy by penalizing divergence between its inferred axiomatic world model and the true physics world. RLAWM probes physics reasoning at two hierarchical levels (modeling and solving), each scored by a physically grounded reward. The modeling reward targets abstract knowledge formulation, evaluating whether the model captures the true axiomatic world before attempting calculation. The solving reward targets instantiated knowledge reasoning, checking whether the model's concrete derivation remains consistent with the axiomatic world. This two-level structure separates what world the model believes it is reasoning in from how it reasons within that world. The framework operates exclusively during training, with no extra cost during inference. RLAWM improves performance by up to +12.2 points on PhysReason and +5.2 on PHYSICS over the baseline. It also generalizes zero-shot across difficulty, source, and domain, with a 2.45x gain and +10 points over the strongest baseline on olympiad-level physics and math.

View arXiv page View PDF Project page Add to collection

Get this paper in your agent:

hf papers read 2610.24519

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.24519 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.24519 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.24519 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co