Paper Citation Record · LEDGER

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization

As of 10 August 2026, this Paper Citation Record lists 25 of 25 outbound references and 0 inbound Pith citation observations for arXiv:2607.19806.

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

pith.paper-citation-record.v1
2607.19806 v1

Coverage vector

measured 25 of 25 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-01T11:43:19.047698Z

measured 25 of 25 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-10T06:31:04.303077+00:00

measured 0 of 0 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links

measured 0 of 1 external citation measurements

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

Source: cited_works

Reference resolution

25 of 25 outbound references displayed

verified exact0
verified fuzzy0
unresolved25
parse uncertain0
malformed identifier0
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation 99deca91-7b85-41c4-8faf-95decc3e4270 · outbound

This paper cites Steering Externalities: Benign Activation Steering Unintentionally Increases Jailbreak Risk for Large Language Models , journal =.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Steering Externalities: Benign Activation Steering Unintentionally Increases Jailbreak Risk for Large Language Models , journal =

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:15.960024Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 09401b0a-4a7f-490c-9447-3f0163bec7f0 · outbound

This paper cites 2024 , eprint=.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization 2024 , eprint=

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:16.115569Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4c0db65e-41d1-481a-8dfc-df3f203e2315 · outbound

This paper cites an unresolved cited work.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Unresolved cited work

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:16.278839Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5f08ff9a-6eea-4a72-9a46-2aa857c5af8a · outbound

This paper cites Steering Llama 2 via Contrastive Activation Addition , booktitle =.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Steering Llama 2 via Contrastive Activation Addition , booktitle =

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:16.474472Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5bf45753-3663-4117-bdd1-fdaac4927fac · outbound

This paper cites Lee and Inkit Padhi and Karthikeyan Natesan Ramamurthy and Erik Miehling and Pierre L.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Lee and Inkit Padhi and Karthikeyan Natesan Ramamurthy and Erik Miehling and Pierre L

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:16.587586Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8e5df596-483a-4b27-b4a9-15983d98c6e7 · outbound

This paper cites an unresolved cited work.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Unresolved cited work

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:16.685080Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 236e916f-83e7-49c0-b9e7-0b11acc3b86f · outbound

This paper cites OR-Bench: An Over-Refusal Benchmark for Large Language Models , booktitle =.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization OR-Bench: An Over-Refusal Benchmark for Large Language Models , booktitle =

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:16.832483Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7045bedc-6f6b-4cd6-a549-1fc689ef6312 · outbound

This paper cites FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoning.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoning

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:16.951980Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 32f73a7f-cf25-462c-b8bf-409f7a67af82 · outbound

This paper cites Jailbroken: How Does.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Jailbroken: How Does

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:17.032879Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 08bbfd56-3e10-4e64-8942-7f303242e748 · outbound

This paper cites Universal and Transferable Adversarial Attacks on Aligned Language Models.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:17.066719Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0e58fd2c-b635-41f3-b91c-024ddd608810 · outbound

This paper cites Forsyth and Dan Hendrycks , editor =.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Forsyth and Dan Hendrycks , editor =

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:17.162942Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f44198c4-c4ec-411f-bb14-8f7d40af1b58 · outbound

This paper cites Pappas and Florian Tram.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Pappas and Florian Tram

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:17.326231Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d0b21cd5-692e-4eba-9600-9f1defaa4ffd · outbound

This paper cites Toy Models of Superposition.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Toy Models of Superposition

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:17.445504Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 84cbdf0d-6807-4a8b-b8fc-b6bec21c9d78 · outbound

This paper cites write code.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization write code

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:17.578787Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8169b8f2-0ecd-4afd-81f2-978db45dcd01 · outbound

This paper cites Representation Engineering: A Top-Down Approach to AI Transparency.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Representation Engineering: A Top-Down Approach to AI Transparency

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:17.692443Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7aec67f2-9847-47ef-923f-dc07266337c0 · outbound

This paper cites Pappas and Eric Wong , title =.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Pappas and Eric Wong , title =

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:17.814760Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b18c0a80-446a-448a-9906-bc4a6dcadaac · outbound

This paper cites Anderson and Yaron Singer and Amin Karbasi , editor =.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Anderson and Yaron Singer and Amin Karbasi , editor =

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:17.949252Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 336babb2-bba3-48ce-b99a-59b8214960a5 · outbound

This paper cites The Thirteenth International Conference on Learning Representations,.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization The Thirteenth International Conference on Learning Representations,

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:18.087080Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 815b3bc2-16cf-4169-8cf1-ae1955e14b72 · outbound

This paper cites Lepori and Lucas Dixon , editor =.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Lepori and Lucas Dixon , editor =

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:18.252718Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f2115387-9a7f-47cd-a74e-4ba1ae071e09 · outbound

This paper cites Galichin and Alexey Dontsov and Oleg Y.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Galichin and Alexey Dontsov and Oleg Y

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:18.417280Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4399139-30bf-4db7-b7ff-b1acadd25547 · outbound

This paper cites Qwen2.5 Technical Report.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Qwen2.5 Technical Report

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:18.585698Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c4b0a056-b9b1-43f0-8991-f34c6325681b · outbound

This paper cites The Llama 3 Herd of Models.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization The Llama 3 Herd of Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:18.725565Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 759ef286-08d1-4acc-8a1a-b25fccc9608d · outbound

This paper cites Advances in Neural Information Processing Systems , volume=.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Advances in Neural Information Processing Systems , volume=

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:18.851358Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bc0c0a92-8040-4b5c-8518-ab3802dde228 · outbound

This paper cites Refusal in Language Models Is Mediated by a Single Direction , url =.

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Refusal in Language Models Is Mediated by a Single Direction , url =

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:18.917386Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a832f09a-9074-4b6f-9b6d-676b4e5ee5a0 · outbound

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

OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization Extending Activation Steering to Broad Skills and Multiple Behaviours

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-01T11:43:19.047698Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

No inbound Pith citation observations are available.