Paper Citation Record · LEDGER

Transformers Use Causal World Models in Maze-Solving Tasks

As of 12 August 2026, this Paper Citation Record lists 25 of 25 outbound references and 3 inbound Pith citation observations for arXiv:2412.11867.

A citation records a reference. It does not transfer a finding from one paper to another.

pith.paper-citation-record.v1
2412.11867 v2

Coverage vector

measured 25 of 25 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-11T14:35:02.320784Z

measured 28 of 28 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-12T06:34:41.77262+00:00

measured 3 of 3 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-06-30T07:29:24.656530Z

measured 0 of 1 external citation measurements

A source-named dated measurement, never combined with another source.

Source: arxiv_reference, observed 2026-06-30T07:34:21.626740Z

Reference resolution

25 of 25 outbound references displayed

verified exact0
verified fuzzy7
unresolved16
parse uncertain0
malformed identifier1
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation ad6820a7-19e1-403c-b153-3420259e9e13 · outbound

This paper cites Understanding intermediate layers using linear classifier probes.

Transformers Use Causal World Models in Maze-Solving Tasks Understanding intermediate layers using linear classifier probes

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.214193Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fb8e1ec8-914d-4e19-819b-4f47327e8ea7 · outbound

This paper cites an unresolved cited work.

Transformers Use Causal World Models in Maze-Solving Tasks Unresolved cited work

Reference 2

Resolution: malformed identifier; raw_fallback, observed 2026-08-11T14:35:02.625767Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 0fcc66e0-6d0c-4c5f-8663-72ea7388104c · outbound

This paper cites Sparse Autoencoders Find Highly Interpretable Features in Language Models.

Transformers Use Causal World Models in Maze-Solving Tasks Sparse Autoencoders Find Highly Interpretable Features in Language Models

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.228788Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6e600833-0be1-47ec-ae0b-6e41b528d5df · outbound

This paper cites 1- and 3-back.

Transformers Use Causal World Models in Maze-Solving Tasks 1- and 3-back

Reference 6

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T14:35:02.612977Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 2fbfb1a6-0ec6-43c2-852e-4c3c43cd9f9a · outbound

This paper cites Evidence of Learned Look-Ahead in a Chess-Playing Neural Network.

Transformers Use Causal World Models in Maze-Solving Tasks Evidence of Learned Look-Ahead in a Chess-Playing Neural Network

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.251568Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 74d95881-dbef-4728-9b18-037837a381a6 · outbound

This paper cites Anthropic circuits Updates - January 2024, Jan- uary.

Transformers Use Causal World Models in Maze-Solving Tasks Anthropic circuits Updates - January 2024, Jan- uary

Reference 10

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T14:35:02.665287Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation b86b0fd5-d3c7-4d0a-83af-ce31c2c8ed5a · outbound

This paper cites Emergent World Models and Latent Variable Estimation in Chess-Playing Language Models.

Transformers Use Causal World Models in Maze-Solving Tasks Emergent World Models and Latent Variable Estimation in Chess-Playing Language Models

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.260686Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fe097f5d-d18e-4551-a47e-89f0af4336ac · outbound

This paper cites Measuring Progress in Dictionary Learning for Language Model Interpretability with Board Game Models.

Transformers Use Causal World Models in Maze-Solving Tasks Measuring Progress in Dictionary Learning for Language Model Interpretability with Board Game Models

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.265630Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ffa14660-9086-4895-a091-95d0c089fc04 · outbound

This paper cites Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task.

Transformers Use Causal World Models in Maze-Solving Tasks Emergent World Representations: Exploring a Sequence Model Trained on a Synthetic Task

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.270102Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d7358571-6f56-4511-93e4-2986f1dc8bc3 · outbound

This paper cites Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla.

Transformers Use Causal World Models in Maze-Solving Tasks Does Circuit Analysis Interpretability Scale? Evidence from Multiple Choice Capabilities in Chinchilla

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.274512Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ce172b65-e52a-4973-9810-58cf95fd697f · outbound

This paper cites A Philosophical Introduction to Language Models - Part II: The Way Forward.

Transformers Use Causal World Models in Maze-Solving Tasks A Philosophical Introduction to Language Models - Part II: The Way Forward

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.278670Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 860adedc-0d51-4d48-b774-b659bcaf6d71 · outbound

This paper cites In-context Learning and Induction Heads.

Transformers Use Causal World Models in Maze-Solving Tasks In-context Learning and Induction Heads

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.282864Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fb3f64f0-6dc3-4956-ac65-ca8c7665e917 · outbound

This paper cites Future Lens: Anticipating Subsequent Tokens from a Single Hidden State.

Transformers Use Causal World Models in Maze-Solving Tasks Future Lens: Anticipating Subsequent Tokens from a Single Hidden State

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.287451Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f57ce243-8517-4410-aa13-d050866bb63f · outbound

This paper cites Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networks.

Transformers Use Causal World Models in Maze-Solving Tasks Toward Transparent AI: A Survey on Interpreting the Inner Structures of Deep Neural Networks

Reference 18

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T14:35:02.652386Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation abb4c6b7-46f6-45ca-9ccb-059987e141d5 · outbound

This paper cites Sparse Relational Reasoning with Object-Centric Representations.

Transformers Use Causal World Models in Maze-Solving Tasks Sparse Relational Reasoning with Object-Centric Representations

Reference 19

Resolution: metadata mismatch; local_arxiv, observed 2026-08-11T14:35:02.361877Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 49894076-1b22-4827-b887-1abb20e781a6 · outbound

This paper cites 16 Published as a conference paper at ICLR 2025 0 1 2 3 4 Maze X 4 3 2 1 0Maze Y 95% (1798,.

Transformers Use Causal World Models in Maze-Solving Tasks 16 Published as a conference paper at ICLR 2025 0 1 2 3 4 Maze X 4 3 2 1 0Maze Y 95% (1798,

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T14:35:02.600405Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 8cf25f77-76e5-4a1c-b50f-a9f95ebb3eb5 · outbound

This paper cites generic edge.

Transformers Use Causal World Models in Maze-Solving Tasks generic edge

Reference 25

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T14:35:02.573839Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 6cd74cfd-3c54-44fc-97c0-c9bed4e904db · outbound

This paper cites an unresolved cited work.

Transformers Use Causal World Models in Maze-Solving Tasks Unresolved cited work

Reference 1619

Resolution: unresolved; raw_fallback, observed 2026-08-11T14:35:02.587225Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 4cb9fc3c-d98d-4eaa-8267-4c3a2d95d717 · outbound

This paper cites Eliciting Latent Predictions from Transformers with the Tuned Lens.

Transformers Use Causal World Models in Maze-Solving Tasks Eliciting Latent Predictions from Transformers with the Tuned Lens

Reference 2016

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.219578Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0316d28b-0e31-4bc5-86fb-a7e79b944b80 · outbound

This paper cites Train Val.

Transformers Use Causal World Models in Maze-Solving Tasks Train Val

Reference 2017

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T14:35:02.638475Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation c8627dc5-4146-435b-83a8-1f641af29e63 · outbound

This paper cites A Configurable Library for Generating and Manipulating Maze Datasets.

Transformers Use Causal World Models in Maze-Solving Tasks A Configurable Library for Generating and Manipulating Maze Datasets

Reference 2019

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.247220Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation cfddd630-c145-45ea-8d31-0c8dac448b06 · outbound

This paper cites Toy Models of Superposition.

Transformers Use Causal World Models in Maze-Solving Tasks Toy Models of Superposition

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.233057Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 44d25516-9f58-4091-bed9-8d8c97c6f7fe · outbound

This paper cites Dictionary Learning Improves Patch-Free Circuit Discovery in Mechanistic Interpretability: A Case Study on Othello-GPT.

Transformers Use Causal World Models in Maze-Solving Tasks Dictionary Learning Improves Patch-Free Circuit Discovery in Mechanistic Interpretability: A Case Study on Othello-GPT

Reference 2022

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.237878Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8c358437-31b3-4607-8bbc-1ad78d6c1e84 · outbound

This paper cites A Mechanistic Analysis of a Transformer Trained on a Symbolic Multi-Step Reasoning Task.

Transformers Use Causal World Models in Maze-Solving Tasks A Mechanistic Analysis of a Transformer Trained on a Symbolic Multi-Step Reasoning Task

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-11T14:35:02.224040Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 580dea35-582c-4dfc-84fb-23ca7e1bf622 · outbound

This paper cites A structural probe for finding syntax in word representa- tions.

Transformers Use Causal World Models in Maze-Solving Tasks A structural probe for finding syntax in word representa- tions

Reference 2024

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T14:35:02.677759Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Pith citing papers

Observation 1245c3f9-0ba0-405f-8c8d-fdf6f0a6fbe9 · inbound

A Co-Evolutionary Theory of Human-AI Coexistence: Mutualism, Governance, and Dynamics in Complex Societies cites this paper.

A Co-Evolutionary Theory of Human-AI Coexistence: Mutualism, Governance, and Dynamics in Complex Societies Transformers Use Causal World Models in Maze-Solving Tasks

Reference 28

Resolution: verified exact; arxiv_id, observed 2026-05-11T20:16:08.936468Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 639df370-b7d8-4661-b2fc-7e397cfb41b7 · inbound

Interaction Locality in Hierarchical Recursive Reasoning cites this paper.

Interaction Locality in Hierarchical Recursive Reasoning Transformers Use Causal World Models in Maze-Solving Tasks

Reference 8

Resolution: verified exact; arxiv_id, observed 2026-05-21T05:04:37.210641Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.

Observation 701e4781-67ff-4f63-b6ea-d7d59ededb3c · inbound

Hierarchical Experimentalist Agents cites this paper.

Hierarchical Experimentalist Agents Transformers Use Causal World Models in Maze-Solving Tasks

Reference 4

Resolution: verified exact; arxiv_id, observed 2026-06-30T07:34:21.628437Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-12T06:34:41.77262+00:00.