Paper Citation Record · LEDGER

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms

As of 11 August 2026, this Paper Citation Record lists 18 of 18 outbound references and 1 inbound Pith citation observation for arXiv:2505.20322.

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

pith.paper-citation-record.v1
2505.20322 v2

Coverage vector

measured 18 of 18 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-07T14:38:54.257012Z

measured 19 of 19 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-11T06:34:44.6726+00:00

measured 1 of 1 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-05T14:22:30.900186Z

measured 0 of 1 external citation measurements

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

Source: pith, observed 2026-08-05T14:22:31.231563Z

Reference resolution

18 of 18 outbound references displayed

verified exact1
verified fuzzy2
unresolved14
parse uncertain0
malformed identifier0
metadata mismatch1

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 87131ce5-1219-4150-87d2-dc40d17d6c30 · outbound

This paper cites LLMSteer: Improving Long-Context LLM Inference by Steering Attention on Reused Contexts.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms LLMSteer: Improving Long-Context LLM Inference by Steering Attention on Reused Contexts

Reference 4

Resolution: metadata mismatch; local_arxiv, observed 2026-08-07T14:38:55.058475Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation df1f811f-fc1a-46fe-89ab-5211f90827b3 · outbound

This paper cites an unresolved cited work.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Unresolved cited work

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:52.514297Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 62e80380-3ae3-44f0-9891-1096766e3240 · outbound

This paper cites OpenAI o1 System Card.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms OpenAI o1 System Card

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:52.606708Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d223a267-db34-414c-9017-a101f0fa270a · outbound

This paper cites Gemma: Open Models Based on Gemini Research and Technology.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Gemma: Open Models Based on Gemini Research and Technology

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:52.836126Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 227f0eb8-5672-4ab5-87b1-ea50cf6b7be0 · outbound

This paper cites Controlling Large Language Model Agents with Entropic Activation Steering.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Controlling Large Language Model Agents with Entropic Activation Steering

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:52.934039Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 6d85ac70-dc1e-4d25-939c-096cd5a8fb31 · outbound

This paper cites A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms A Systematic Survey of Prompt Engineering in Large Language Models: Techniques and Applications

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:53.027749Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d9778b3d-6723-4ebf-9dc5-254bade878d7 · outbound

This paper cites Multi-property Steering of Large Language Models with Dynamic Activation Composition.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Multi-property Steering of Large Language Models with Dynamic Activation Composition

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:53.145439Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3731968c-4dc2-4ff6-b8df-9460f1f6b739 · outbound

This paper cites Alexander Matt Turner, Lisa Thiergart, Gavin Leech, David Udell, Juan J Vazquez, Ulisse Mini, and Monte MacDiarmid.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Alexander Matt Turner, Lisa Thiergart, Gavin Leech, David Udell, Juan J Vazquez, Ulisse Mini, and Monte MacDiarmid

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:38:55.716620Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 65d5f75e-9d4f-4a21-bf3a-12c4e2c5b3c3 · outbound

This paper cites Extending Activation Steering to Broad Skills and Multiple Behaviours.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Extending Activation Steering to Broad Skills and Multiple Behaviours

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:53.484364Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 987db901-2058-4dc1-846a-95728c3f3a52 · outbound

This paper cites Trading Inference-Time Compute for Adversarial Robustness.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Trading Inference-Time Compute for Adversarial Robustness

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:53.777359Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d9bac481-c22f-482d-ae1d-883d84865606 · outbound

This paper cites A Survey of Large Language Models.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms A Survey of Large Language Models

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:53.960690Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56f85661-cf8b-4a83-bac5-a082ce149ce4 · outbound

This paper cites Conversely, our research focuses on improv- ing detoxification efficacy while preserving robust general performance to ensure practical model util- ity.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Conversely, our research focuses on improv- ing detoxification efficacy while preserving robust general performance to ensure practical model util- ity

Reference 17

Resolution: verified fuzzy; raw_fallback, observed 2026-08-07T14:38:55.415082Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation b9181426-b2ee-4cbc-b5c0-4d8e5329c065 · outbound

This paper cites To maximize their effectiveness, we concatenate these prompts at var- ious positions, including the input prefix, input suffix, and output prefix.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms To maximize their effectiveness, we concatenate these prompts at var- ious positions, including the input prefix, input suffix, and output prefix

Reference 18

Resolution: verified exact; raw_fallback, observed 2026-08-07T14:38:54.549573Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.

Observation 321a4499-5630-4381-97bd-fa3ec0eeb4f0 · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Training Verifiers to Solve Math Word Problems

Reference 2021

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:52.171316Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 57092afc-9ba7-4e60-b898-bade1c5ef9c0 · outbound

This paper cites EasyEdit: An Easy-to-use Knowledge Editing Framework for Large Language Models.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms EasyEdit: An Easy-to-use Knowledge Editing Framework for Large Language Models

Reference 2023

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:53.618016Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3d775816-7bcf-4e05-8ed6-7fdbe7c05813 · outbound

This paper cites Improving Steering Vectors by Targeting Sparse Autoencoder Features.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Improving Steering Vectors by Targeting Sparse Autoencoder Features

Reference 2024

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:52.048185Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation eef794d5-2750-44eb-813a-1ab3f050ded6 · outbound

This paper cites The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Tasks.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms The Danger of Overthinking: Examining the Reasoning-Action Dilemma in Agentic Tasks

Reference 2025

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:52.251169Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 39ea6fbb-1ff8-4eb9-a5af-a13a7f6900a9 · outbound

This paper cites Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models.

Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms Sparse Feature Circuits: Discovering and Editing Interpretable Causal Graphs in Language Models

Reference 8098

Resolution: unresolved; no resolver link, observed 2026-08-07T14:38:52.676701Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation f855ebe7-01e3-4b09-8838-7c16e69a614a · inbound

When Models Refuse: Political Steerability and Feature Richness as Measures of Ideological Depth cites this paper.

When Models Refuse: Political Steerability and Feature Richness as Measures of Ideological Depth Beyond Prompt Engineering: Robust Behavior Control in LLMs via Steering Target Atoms

Reference 27

Resolution: verified exact; local_arxiv, observed 2026-08-05T14:22:31.238594Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.