Paper Citation Record · LEDGER

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal

As of 8 August 2026, this Paper Citation Record lists 27 of 27 outbound references and 2 inbound Pith citation observations for arXiv:2602.04521.

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

pith.paper-citation-record.v1
2602.04521 v2

Coverage vector

measured 27 of 27 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-03T04:40:20.490400Z

measured 29 of 29 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-08T06:32:00.761636+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-07-07T15:35:54.665268Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-07-07T15:43:53.876618Z

Reference resolution

27 of 27 outbound references displayed

verified exact0
verified fuzzy0
unresolved26
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation ead5c0cc-f144-4236-bed0-fdc800558b2f · outbound

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

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Refusal in Language Models Is Mediated by a Single Direction

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:19.755615Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e5dadb6f-d8b8-42da-b418-0d5fb82a5586 · outbound

This paper cites Lee, Inkit Padhi, Karthikeyan Natesan Ramamurthy, Erik Miehling, Pierre Dognin, Manish Nagireddy, and Amit Dhurandhar.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Lee, Inkit Padhi, Karthikeyan Natesan Ramamurthy, Erik Miehling, Pierre Dognin, Manish Nagireddy, and Amit Dhurandhar

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:19.848081Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ec41b4c3-dff8-48e2-b5c7-3d95beca7503 · outbound

This paper cites Measuring massive multitask language understanding.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Measuring massive multitask language understanding

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:19.957594Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e3abd516-894b-4daa-9354-a91e6a904d0d · outbound

This paper cites Training verifiers to solve math word problems, 2021.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Training verifiers to solve math word problems, 2021

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.041635Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 99ab51e0-3848-44b1-95eb-7a933152af0c · outbound

This paper cites SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal SORRY-Bench: Systematically Evaluating Large Language Model Safety Refusal

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.100340Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2b2a3439-97bf-4a83-b34a-a78bf6a3a095 · outbound

This paper cites Steering Language Models With Activation Engineering.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Steering Language Models With Activation Engineering

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.153002Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2e7c9acc-362f-42b2-bcf1-3e495e1a2389 · outbound

This paper cites Extracting latent steering vectors from pretrained language models.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Extracting latent steering vectors from pretrained language models

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.228228Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b0aa0ea5-da8c-4580-901b-d712361fb910 · outbound

This paper cites Beyond prompt engineering: Robust behavior control in LLMs via steering target atoms.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Beyond prompt engineering: Robust behavior control in LLMs via steering target atoms

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.285537Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 32ff9ec6-5a25-4f89-a106-647241bb6a51 · outbound

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

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Representation Engineering: A Top-Down Approach to AI Transparency

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.362627Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0d26188a-2ad3-435c-9ee9-a099159d02cd · outbound

This paper cites Steering llama 2 via contrastive activation addition.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Steering llama 2 via contrastive activation addition

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.415647Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5cfd39f2-998b-4ae6-97ce-91a7908ac32c · outbound

This paper cites Inference-time intervention: Eliciting truthful answers from a language model.Advances in Neural Information Processing Systems, 36:41451– 41530, 2023.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Inference-time intervention: Eliciting truthful answers from a language model.Advances in Neural Information Processing Systems, 36:41451– 41530, 2023

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.450158Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4d1418b-b991-461a-b4de-98f607b7d385 · outbound

This paper cites Editing models with task arithmetic.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Editing models with task arithmetic

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.452736Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f769f365-be45-4a80-9e71-0d8941899e14 · outbound

This paper cites Ties-merging: Resolving interference when merging models.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Ties-merging: Resolving interference when merging models

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.455244Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 58cba416-8975-40fa-9bfe-31c6b9843f3a · outbound

This paper cites Localizing task information for improved model merging and compression.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Localizing task information for improved model merging and compression

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.457561Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c11ca62f-5248-4274-9c9e-33ff3b4dbe6f · outbound

This paper cites Model breadcrumbs: Scaling multi-task model merging with sparse masks.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Model breadcrumbs: Scaling multi-task model merging with sparse masks

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.460048Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 511d4809-31a5-4750-92a6-1e3a00b1c3b0 · outbound

This paper cites Lines: Post-training layer scaling prevents forgetting and enhances model merging.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Lines: Post-training layer scaling prevents forgetting and enhances model merging

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.462364Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c7e424a6-c0a2-4b88-9c30-a05c1f1fd7f4 · outbound

This paper cites Steering language models with weight arithmetic, 2025.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Steering language models with weight arithmetic, 2025

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.464697Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 300f99d0-8169-4bda-b329-67a83ba82979 · outbound

This paper cites Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.466986Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 058e6a72-df2f-4bee-89ed-73b588ec0ef2 · outbound

This paper cites Axiomatic attribution for deep networks, 2017.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Axiomatic attribution for deep networks, 2017

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.470086Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation bf9de0f3-b618-48de-ad93-56013db5f24d · outbound

This paper cites an unresolved cited work.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Unresolved cited work

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.472442Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 046491f1-6cfd-4782-bd49-54dc9e36c340 · outbound

This paper cites an unresolved cited work.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Unresolved cited work

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.475201Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c2f21251-6ad1-4c18-9987-534e915964b4 · outbound

This paper cites A framework for few-shot language model evaluation, September 2021.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal A framework for few-shot language model evaluation, September 2021

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.477768Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2cf6bc52-ff14-47fc-8929-36a07ac37030 · outbound

This paper cites Activation Addition.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Activation Addition

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.480024Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8f0599c-7386-4719-9f5c-920348eb840a · outbound

This paper cites If the classifier predicts a refusal (indicated by is_refusal = 1), the judgment is finalized immediately.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal If the classifier predicts a refusal (indicated by is_refusal = 1), the judgment is finalized immediately

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.482487Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 88269c9f-2c5e-461e-83f8-f946f2195226 · outbound

This paper cites The judge is tasked with a binary classification based on a strict rubric.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal The judge is tasked with a binary classification based on a strict rubric

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.484950Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7bec0fc8-ebd2-42f1-9356-888b227e6ea1 · outbound

This paper cites an unresolved cited work.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Unresolved cited work

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-03T04:40:20.487467Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2a5d3dda-5c7f-4ee2-9077-100fa9e635e6 · outbound

This paper cites Return only the single word: answered | refused.

$C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal Return only the single word: answered | refused

Reference 27

Resolution: malformed identifier; no resolver link, observed 2026-08-03T04:40:20.490400Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Pith citing papers

Observation 531cf99b-422f-489f-970b-958087864a97 · inbound

Forgetting That Sticks: Quantization-Permanent Unlearning via Circuit Attribution cites this paper.

Forgetting That Sticks: Quantization-Permanent Unlearning via Circuit Attribution $C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal

Reference 17

Resolution: verified exact; arxiv_id, observed 2026-07-09T02:20:28.526184Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.

Observation 8d959f8d-bf88-4b6d-9326-b1334924beb8 · inbound

Faithfulness to Refusal: A Causal Audit of Neuron Selectors cites this paper.

Faithfulness to Refusal: A Causal Audit of Neuron Selectors $C$-$\Delta\Theta$: Circuit-Restricted Weight Arithmetic for Selective Refusal

Reference 31

Resolution: verified exact; arxiv_id, observed 2026-07-09T02:20:28.526184Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.