# Passive learning of active causal strategies in agents and language models

arXiv: https://arxiv.org/abs/2305.16183 · HTML: https://mlsanity.com/paper/2305.16183/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- Placed in Medical and scientific domain agents, one of 16 directions in LLM agents and tool use.
- Builds on 52 of its 77 references that are in the map.
- First posted on arXiv 25 May 2023.

## Placed in

- LLM agents and tool use / [Medical and scientific domain agents](https://mlsanity.com/agents/medical-and-scientific-domain-agents/)

## Builds on (most cited first, 30 of 52)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [Transformer-XL: Attentive Language Models Beyond a Fixed-Length Context](https://mlsanity.com/paper/1901.02860/) (2019)
- [Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Them](https://mlsanity.com/paper/2210.09261/) (2022)
- [Decision Transformer: Reinforcement Learning via Sequence Modeling](https://mlsanity.com/paper/2106.01345/) (2021)
- [Off-Policy Deep Reinforcement Learning without Exploration](https://mlsanity.com/paper/1812.02900/) (2018)
- [LIMA: Less Is More for Alignment](https://mlsanity.com/paper/2305.11206/) (2023)
- [Show Your Work: Scratchpads for Intermediate Computation with Language Models](https://mlsanity.com/paper/2112.00114/) (2021)
- [Stabilizing Off-Policy Q-Learning via Bootstrapping Error Reduction](https://mlsanity.com/paper/1906.00949/) (2019)
- [Distilling Step-by-Step! Outperforming Larger Language Models with Less Training Data and Smaller Model Sizes](https://mlsanity.com/paper/2305.02301/) (2023)
- [Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task](https://mlsanity.com/paper/2210.13382/) (2022)
- [Causal Reasoning and Large Language Models: Opening a New Frontier for Causality](https://mlsanity.com/paper/2305.00050/) (2023)
- [Causal Confusion in Imitation Learning](https://mlsanity.com/paper/1905.11979/) (2019)
- [Can language models learn from explanations in context?](https://mlsanity.com/paper/2204.02329/) (2022)
- [Grounding Large Language Models in Interactive Environments with Online Reinforcement Learning](https://mlsanity.com/paper/2302.02662/) (2023)
- [In-context Reinforcement Learning with Algorithm Distillation](https://mlsanity.com/paper/2210.14215/) (2022)
- [Implicit Representations of Meaning in Neural Language Models](https://mlsanity.com/paper/2106.00737/) (2021)
- [WT5?! Training Text-to-Text Models to Explain their Predictions](https://mlsanity.com/paper/2004.14546/) (2020)
- [Foundations of Structural Causal Models with Cycles and Latent Variables](https://mlsanity.com/paper/1611.06221/) (2016)
- [Language Models as Agent Models](https://mlsanity.com/paper/2212.01681/) (2022)
- [Reinforcement Learning Upside Down: Don't Predict Rewards -- Just Map Them to Actions](https://mlsanity.com/paper/1912.02875/) (2019)
- [Teaching Algorithmic Reasoning via In-context Learning](https://mlsanity.com/paper/2211.09066/) (2022)
- [Training Agents using Upside-Down Reinforcement Learning](https://mlsanity.com/paper/1912.02877/) (2019)

## Built on by (most cited first, 21 of 21)

- [Challenges and Applications of Large Language Models](https://mlsanity.com/paper/2307.10169/) (2023)
- [Bayesian Low-rank Adaptation for Large Language Models](https://mlsanity.com/paper/2308.13111/) (2023)
- [The broader spectrum of in-context learning](https://mlsanity.com/paper/2412.03782/) (2024)
- [From Language to Action: A Review of Large Language Models as Autonomous Agents and Tool Users](https://mlsanity.com/paper/2508.17281/) (2025)
- [ALCM: Autonomous LLM-Augmented Causal Discovery Framework](https://mlsanity.com/paper/2405.01744/) (2024)
- [Latent learning: episodic memory complements parametric learning by enabling flexible reuse of experiences](https://mlsanity.com/paper/2509.16189/) (2025)
- [Are Large Language Models Reliable AI Scientists? Assessing Reverse-Engineering of Black-Box Systems](https://mlsanity.com/paper/2505.17968/) (2025)
- [Can Large Language Models Help Experimental Design for Causal Discovery?](https://mlsanity.com/paper/2503.01139/) (2025)
- [Language Models are Bounded Pragmatic Speakers: Understanding RLHF from a Bayesian Cognitive Modeling Perspective](https://mlsanity.com/paper/2305.17760/) (2023)
- [Amortized Active Causal Induction with Deep Reinforcement Learning](https://mlsanity.com/paper/2405.16718/) (2024)
- [Context Structure Reshapes the Representational Geometry of Language Models](https://mlsanity.com/paper/2601.22364/) (2026)
- [Functional-level Uncertainty Quantification for Calibrated Fine-tuning on LLMs](https://mlsanity.com/paper/2410.06431/) (2024)
- [Normative Conflicts and Shallow AI Alignment](https://mlsanity.com/paper/2506.04679/) (2025)
- [CausaLab: A Scalable Environment for Interactive Causal Discovery Toward AI Scientists](https://mlsanity.com/paper/2605.26029/) (2026)
- [IGDA: Interactive Graph Discovery through Large Language Model Agents](https://mlsanity.com/paper/2502.17189/) (2025)
- [Fodor and Pylyshyn's Legacy: Still No Human-like Systematic Compositionality in Neural Networks](https://mlsanity.com/paper/2506.01820/) (2025)
- [Systems with Switching Causal Relations: A Meta-Causal Perspective](https://mlsanity.com/paper/2410.13054/) (2024)
- [Reasoning about Affordances: Causal and Compositional Reasoning in LLMs](https://mlsanity.com/paper/2502.16606/) (2025)
- [Spatial Reasoning in LLM Game Agents: Impact of Causal Context and Multi-Step Planning](https://mlsanity.com/paper/2607.22732/) (2026)
- [Representation in large language models](https://mlsanity.com/paper/2501.00885/) (2025)
- [Evidence-Type Competition: When Can Interventional Data Teach Language Models Causal Direction?](https://mlsanity.com/paper/2607.29484/) (2026)

## Abstract (arXiv)

What can be learned about causality and experimentation from passive data? This question is salient given recent successes of passively-trained language models in interactive domains such as tool use. Passive learning is inherently limited. However, we show that purely passive learning can in fact allow an agent to learn generalizable strategies for determining and using causal structures, as long as the agent can intervene at test time. We formally illustrate that learning a strategy of first experimenting, then seeking goals, can allow generalization from passive learning in principle. We then show empirically that agents trained via imitation on expert data can indeed generalize at test time to infer and use causal links which are never present in the training data; these agents can also generalize experimentation strategies to novel variable sets never observed in training. We then show that strategies for causal intervention and exploitation can be generalized from passive data even in a more complex environment with high-dimensional observations, with the support of natural language explanations. Explanations can even allow passive learners to generalize out-of-distribution from perfectly-confounded training data. Finally, we show that language models, trained only on passive next-word prediction, can generalize causal intervention strategies from a few-shot prompt containing examples of experimentation, together with explanations and reasoning. These results highlight the surprising power of passive learning of active causal strategies, and may help to understand the behaviors and capabilities of language models.