Generalization of RLVR Using Causal Reasoning as a Testbed

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Lu, Brian, Zhao, Hongyu, Sun, Shuo, Peng, Hao, Ding, Rui, Mei, Hongyuan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910040813731840
author Lu, Brian
Zhao, Hongyu
Sun, Shuo
Peng, Hao
Ding, Rui
Mei, Hongyuan
author_facet Lu, Brian
Zhao, Hongyu
Sun, Shuo
Peng, Hao
Ding, Rui
Mei, Hongyuan
contents Reinforcement learning with verifiable rewards (RLVR) has emerged as a promising paradigm for post-training large language models (LLMs) on complex reasoning tasks. Yet, the conditions under which RLVR yields robust generalization remain underexplored. This paper provides an empirical study of RLVR generalization in the setting of probabilistic inference over causal graphical models. This setting offers two natural axes along which to examine generalization: (i) the level of the probabilistic query -- associational, interventional, or counterfactual -- and (ii) the structural complexity of the query, measured by the size of its relevant subgraph. We construct a dataset of causal graphs and queries spanning these difficulty axes and fine-tune Qwen-2.5-Instruct models using RLVR or supervised fine-tuning (SFT). We vary both the model scale (3B-32B) and the query level included in training. We find that RLVR yields stronger within-level and across-level generalization than SFT, but only for specific combinations of model size and training query level. Further analysis shows that RLVR's effectiveness depends on the model's initial reasoning competence. With sufficient initial competence, RLVR improves an LLM's marginalization strategy and reduces errors in intermediate probability calculations, producing substantial accuracy gains, particularly on more complex queries. These results show that RLVR can improve specific causal reasoning subskills, with its benefits emerging only when the model has sufficient initial competence. Our code and data is available at https://github.com/zhichul/rlcausal.
format Preprint
id arxiv_https___arxiv_org_abs_2512_20760
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Generalization of RLVR Using Causal Reasoning as a Testbed
Lu, Brian
Zhao, Hongyu
Sun, Shuo
Peng, Hao
Ding, Rui
Mei, Hongyuan
Machine Learning
Artificial Intelligence
Computation and Language
Reinforcement learning with verifiable rewards (RLVR) has emerged as a promising paradigm for post-training large language models (LLMs) on complex reasoning tasks. Yet, the conditions under which RLVR yields robust generalization remain underexplored. This paper provides an empirical study of RLVR generalization in the setting of probabilistic inference over causal graphical models. This setting offers two natural axes along which to examine generalization: (i) the level of the probabilistic query -- associational, interventional, or counterfactual -- and (ii) the structural complexity of the query, measured by the size of its relevant subgraph. We construct a dataset of causal graphs and queries spanning these difficulty axes and fine-tune Qwen-2.5-Instruct models using RLVR or supervised fine-tuning (SFT). We vary both the model scale (3B-32B) and the query level included in training. We find that RLVR yields stronger within-level and across-level generalization than SFT, but only for specific combinations of model size and training query level. Further analysis shows that RLVR's effectiveness depends on the model's initial reasoning competence. With sufficient initial competence, RLVR improves an LLM's marginalization strategy and reduces errors in intermediate probability calculations, producing substantial accuracy gains, particularly on more complex queries. These results show that RLVR can improve specific causal reasoning subskills, with its benefits emerging only when the model has sufficient initial competence. Our code and data is available at https://github.com/zhichul/rlcausal.
title Generalization of RLVR Using Causal Reasoning as a Testbed
topic Machine Learning
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2512.20760