paper-with-me

Papers

Uncovering Spontaneous Physics Representations in In-Context Learning

2025-08-17 · Yeongwoo Song, Jaeyong Bae, Dong-Kyum Kim, Hawoong Jeong arxiv

In-context learning (ICL) lets large language models (LLMs) solve new tasks from prompts alone, across an ever-widening range of domains, yet the mechanisms underlying this ability remain poorly understood. Physical systems offer a controlled testbed for this question as they provide experimentally controllable data with structured dynamics grounded in fundamental principles. Here we study the ICL ability of LLMs, focusing on physical reasoning. Using dynamics forecasting as a proxy task, we first show that LLMs forecast physical dynamics in context, with accuracy improving as more history is provided. Analyzing the model's residual stream reveals internal activations that correlate with key physical quantities such as energy. These correlations strengthen gradually with context length, indicating that LLMs spontaneously form representations aligned with physical concepts without any physics-specific supervision. To assess whether these representations contribute to the model's predictions, we introduce a layer-wise gradient-based attribution analysis. We find that, residual directions more strongly correlated with energy also receive greater attribution to numerical predictions. This pattern is not observed for features correlated with directly observed quantities such as displacement, suggesting that the energy-related signal is not merely numerical information copied from the input. Our results broaden ICL analysis to structured physical dynamics and give a mechanistic account of how LLMs organize physical structure in context.

📄 PDF Abstract BibTeX arXiv:2508.12448

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Complexity for deep neural networks and other characteristics of deep feature representations

2020-06-08 · Romuald A. Janik, Przemek Witaszczyk

We define a notion of complexity, which quantifies the nonlinearity of the computation of a neural network, as well as a complementary measure of the effective dimension of feature representations. We investigate these o…

Minimal model of permutation symmetry in unsupervised learning

2019-04-30 · Tianqi Hou, K. Y. Michael Wong, Haiping Huang

Permutation of any two hidden units yields invariant properties in typical deep generative neural networks. This permutation symmetry plays an important role in understanding the computation performance of a broad class …

model

Visuomotor Mechanical Search: Learning to Retrieve Target Objects in Clutter

2020-08-13 · Andrey Kurenkov, Joseph Taglic, Rohun Kulkarni, Marcus Dominguez-Kuhne 외

When searching for objects in cluttered environments, it is often necessary to perform complex interactions in order to move occluding objects out of the way and fully reveal the object of interest and make it graspable.…

Deep Reinforcement LearningObjectReinforcement Learning (RL)Retrieval

On the Use of Self-Supervised Speech Representations in Spontaneous Speech Synthesis

2023-07-11 · Siyang Wang, Gustav Eje Henter, Joakim Gustafson, Éva Székely

Self-supervised learning (SSL) speech representations learned from large amounts of diverse, mixed-quality speech data without transcriptions are gaining ground in many speech technology applications. Prior work has show…

PredictionSelf-Supervised LearningSpeech Synthesistext-to-speech+1

Dynamics of Spontaneous Topic Changes in Next Token Prediction with Self-Attention

2025-01-10 · Mumin Jia, Jairo Diaz-Rodriguez

Human cognition is punctuated by abrupt, spontaneous shifts between topics-driven by emotional, contextual, or associative cues-a phenomenon known as spontaneous thought in neuroscience. In contrast, self-attention based…