paper-with-me

홈 › Papers

Rule Extrapolation in Language Models: A Study of Compositional Generalization on OOD Prompts

2024-09-09 · Anna Mészáros, Szilvia Ujváry, Wieland Brendel, Patrik Reizinger, Ferenc Huszár

LLMs show remarkable emergent abilities, such as inferring concepts from presumably out-of-distribution prompts, known as in-context learning. Though this success is often attributed to the Transformer architecture, our systematic understanding is limited. In complex real-world data sets, even defining what is out-of-distribution is not obvious. To better understand the OOD behaviour of autoregressive LLMs, we focus on formal languages, which are defined by the intersection of rules. We define a new scenario of OOD compositional generalization, termed rule extrapolation. Rule extrapolation describes OOD scenarios, where the prompt violates at least one rule. We evaluate rule extrapolation in formal languages with varying complexity in linear and recurrent architectures, the Transformer, and state space models to understand the architectures' influence on rule extrapolation. We also lay the first stones of a normative theory of rule extrapolation, inspired by the Solomonoff prior in algorithmic information theory.

📄 PDF Abstract BibTeX arXiv:2409.13728

Code (1)

meszarosanna/rule_extrapolation 공식 구현 pytorch

Tasks

In-Context LearningState Space Models

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Position-Wise Feed-Forward Layer 설명 없음
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Similar Papers 제목 키워드 기반

Loop, Think, & Generalize: Implicit Reasoning in Recurrent-Depth Transformers

2026-04-09 · Harsh Kohli, Srinivasan Parthasarathy, Huan Sun, Yuekun Yao arxiv

We study implicit reasoning, i.e. the ability to combine knowledge or rules within a single forward pass. While transformer-based large language models store substantial factual knowledge and rules, they often fail to co…

Im-Promptu: In-Context Composition from Image Prompts

2023-05-26 · NeurIPS 2023 11 · Bhishma Dedhia, Michael Chang, Jake C. Snell, Thomas L. Griffiths 외

Large language models are few-shot learners that can solve diverse tasks from a handful of demonstrations. This implicit understanding of tasks suggests that the attention mechanisms over word tokens may play a role in a…

Image GenerationMeta-Learning

Differentiable Compositional Kernel Learning for Gaussian Processes

2018-06-12 · ICML 2018 7 · Shengyang Sun, Guodong Zhang, Chaoqi Wang, Wenyuan Zeng 외

The generalization properties of Gaussian processes depend heavily on the choice of kernel, and this choice remains a dark art. We present the Neural Kernel Network (NKN), a flexible family of kernels represented by a ne…

Bayesian OptimizationGaussian ProcessesTime SeriesTime Series Analysis+1

Out-of-distribution Tests Reveal Compositionality in Chess Transformers

2025-10-23 · Anna Mészáros, Patrik Reizinger, Ferenc Huszár arxiv

Chess is a canonical example of a task that requires rigorous reasoning and long-term planning. Modern decision Transformers - trained similarly to LLMs - are able to learn competent gameplay, but it is unclear to what e…

Linguistic generalization and compositionality in modern artificial neural networks

2019-03-30 · Marco Baroni

In the last decade, deep artificial neural networks have achieved astounding performance in many natural language processing tasks. Given the high productivity of language, these models must possess effective generalizat…

Systematic Generalization