Paper Citation Record · LEDGER

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests

As of 20 August 2026, this Paper Citation Record lists 62 of 62 outbound references and 2 inbound Pith citation observations for arXiv:2502.06867.

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

pith.paper-citation-record.v1
2502.06867 v1

Coverage vector

measured 62 of 62 reference resolution

Typed states for the displayed outbound observations.

Source: paper_references, paper_reference_links, observed 2026-08-08T19:21:19.261269Z

measured 64 of 64 standing notices

One-hop event checks from named stored sources.

Source: scholarly_work_events, retraction_status_cache, observed 2026-08-20T06:33:59.587034+00:00

measured 2 of 2 inbound itemization

Pith citing papers itemized under the disclosed page cap.

Source: paper_references, paper_reference_links, observed 2026-08-01T14:45:42.876760Z

measured 0 of 1 external citation measurements

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

Source: arxiv_reference, observed 2026-05-22T01:20:51.933250Z

Reference resolution

62 of 62 outbound references displayed

verified exact2
verified fuzzy15
unresolved44
parse uncertain0
malformed identifier1
metadata mismatch0

External citation measurements

No source-named external measurement is stored.

Outbound references

Observation dbe955ae-1107-4c8a-bacb-47a8f4ae682b · outbound

This paper cites Benchmark Early and Red Team Often: A Framework for Assessing and Managing Dual-Use Hazards of AI Foundation Models.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Benchmark Early and Red Team Often: A Framework for Assessing and Managing Dual-Use Hazards of AI Foundation Models

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:18.977393Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9ed0c418-1c4c-44b4-946b-b29d5bc43789 · outbound

This paper cites Adversaries Can Misuse Combinations of Safe Models.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Adversaries Can Misuse Combinations of Safe Models

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:18.980851Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 039ac423-8e6e-4165-a7eb-c26dca054a9e · outbound

This paper cites Hazards from Increasingly Accessible Fine-Tuning of Downloadable Foundation Models.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Hazards from Increasingly Accessible Fine-Tuning of Downloadable Foundation Models

Reference 3

Resolution: verified exact; local_arxiv, observed 2026-08-08T19:21:19.435175Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 8d03ed14-e472-4add-a81e-40852ce8cfd0 · outbound

This paper cites Defense Priorities in the Open-Source AI Debate: A Preliminary Assessment.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Defense Priorities in the Open-Source AI Debate: A Preliminary Assessment

Reference 4

Resolution: verified exact; local_arxiv, observed 2026-08-08T19:21:19.424997Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation de131ff2-0e55-4cdb-bd0f-12581a6b08dc · outbound

This paper cites Control Risk for Potential Misuse of Artificial Intelligence in Science.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Control Risk for Potential Misuse of Artificial Intelligence in Science

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:18.989739Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e4e6a693-7264-4b03-92ed-67716ab05ef7 · outbound

This paper cites HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:18.992928Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 733f9edd-a103-4b37-b22e-631eacf22603 · outbound

This paper cites An Overview of Catastrophic AI Risks.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests An Overview of Catastrophic AI Risks

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:18.996829Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b4b1312c-51f3-43cb-b750-05173ca6efc2 · outbound

This paper cites Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.000586Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2684cff5-a8d2-41ca-b436-99accad025da · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 9

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.742322Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation f617b400-b5a7-4967-a481-894f907de102 · outbound

This paper cites Emergent autonomous scientific research capabilities of large language models.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Emergent autonomous scientific research capabilities of large language models

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.007290Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 39427626-96cd-45f7-b29e-4926c68eb168 · outbound

This paper cites Emergent Language: A Survey and Taxonomy.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Emergent Language: A Survey and Taxonomy

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.103143Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ff05affc-bf86-4205-a8b7-80620f1f1db2 · outbound

This paper cites A., Caliskan, A., Liyanage, S., & Banaji, M.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests A., Caliskan, A., Liyanage, S., & Banaji, M

Reference 12

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.734559Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation e5039700-dae9-4cc0-926f-3404a875901c · outbound

This paper cites J., Kaplan, D., Ren, Z., Hsu, C.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests J., Kaplan, D., Ren, Z., Hsu, C

Reference 13

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.726864Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation b74659b5-d7e7-420b-83d1-f12092732622 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 14

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.719005Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation f9932c7e-a28c-401f-9fe9-64218e7754a1 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 15

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.711338Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation d1db077f-503e-4eae-9c25-24f001ecb6a0 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 16

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.703674Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 434fe21e-5040-42c6-87bc-a6b898ddc71f · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 17

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.695182Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation e67b5b15-8613-41b6-8f52-7b5cf9eaefc6 · outbound

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

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests OR-Bench: An Over-Refusal Benchmark for Large Language Models

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.126985Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4edab477-976c-45f3-aabe-a4cfe1f36a36 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 19

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.687582Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 99e748b6-a813-4ae1-bd9b-7daf04612e88 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 20

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.680956Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 97659b8a-d5bd-486d-83fd-10fb7c270555 · outbound

This paper cites Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Models.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Models

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.137233Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation edfd4c34-154b-4fac-9aec-73ecd36b53cc · outbound

This paper cites J., & Wick, M.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests J., & Wick, M

Reference 22

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.674298Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 3beb04eb-39ec-44cf-bd17-2bac03f6d8c1 · outbound

This paper cites POROver: Improving Safety and Reducing Overrefusal in Large Language Models with Overgeneration and Preference Optimization.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests POROver: Improving Safety and Reducing Overrefusal in Large Language Models with Overgeneration and Preference Optimization

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.143563Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0503e350-69ab-4040-9474-d413bfc24ec8 · outbound

This paper cites Automatic Pseudo-Harmful Prompt Generation for Evaluating False Refusals in Large Language Models.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Automatic Pseudo-Harmful Prompt Generation for Evaluating False Refusals in Large Language Models

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.146827Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3087a14e-c766-4a59-b8b0-d9a0e8a861d2 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 25

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.667729Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation cc75fb0e-a32f-46d6-bedf-0ae9e2acca5e · outbound

This paper cites Can large language models democratize access to dual-use biotechnology?.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Can large language models democratize access to dual-use biotechnology?

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.153433Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5e4ce984-71d0-4a24-8fc2-354289a40960 · outbound

This paper cites The Reality of AI and Biorisk.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests The Reality of AI and Biorisk

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.157071Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f98ae51-7924-4007-ae1b-9f5ec35e1d84 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 28

Resolution: malformed identifier; no resolver link, observed 2026-08-08T19:21:19.160347Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 20c8e7c3-5664-4860-ba33-0cdb2596ac2e · outbound

This paper cites H., Rocha, R., Cordova, K., Specter, M., & Esvelt, K.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests H., Rocha, R., Cordova, K., Specter, M., & Esvelt, K

Reference 29

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.659994Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 534a0b96-fa5f-46af-8d1b-5d2c769967a5 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 30

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.652162Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation db6dd4e7-ed62-4e45-9684-f88c32225e1a · outbound

This paper cites Will releasing the weights of future large language models grant widespread access to pandemic agents?.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Will releasing the weights of future large language models grant widespread access to pandemic agents?

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.167805Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 847b202b-b044-492c-8ded-eb0f5cdf655b · outbound

This paper cites Prioritizing High-Consequence Biological Capabilities in Evaluations of Artificial Intelligence Models.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Prioritizing High-Consequence Biological Capabilities in Evaluations of Artificial Intelligence Models

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.170579Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation c55da87b-582f-40a9-b76c-57885508d326 · outbound

This paper cites Towards Responsible Governance of Biological Design Tools.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Towards Responsible Governance of Biological Design Tools

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.173371Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e6dc54b5-35a0-4c2a-b428-bdb0d36abb73 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 34

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.644747Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 7e635a9a-e18e-4af6-af7f-45a4222334c0 · outbound

This paper cites LAB-Bench: Measuring Capabilities of Language Models for Biology Research.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests LAB-Bench: Measuring Capabilities of Language Models for Biology Research

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.179354Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 02a83956-c979-4585-aca1-ed9482a10ff6 · outbound

This paper cites T., Martin, F.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests T., Martin, F

Reference 36

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.637172Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 87873364-4b4c-49f2-bda4-3e645e0bd11b · outbound

This paper cites Are large language models superhuman chemists?.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Are large language models superhuman chemists?

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.185989Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f7cc2bbf-a230-4360-93f0-a9c784beb9aa · outbound

This paper cites M., Ai, Q., Al-Feghali, A., Badhwar, S., Bocarsly, J.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests M., Ai, Q., Al-Feghali, A., Badhwar, S., Bocarsly, J

Reference 38

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.629717Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 8e9c0b36-8504-423d-8708-66dbd787a183 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 39

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.622247Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 8f4b8a26-4641-4beb-8258-e0c66b8b89ee · outbound

This paper cites ChemCrow: Augmenting large-language models with chemistry tools.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests ChemCrow: Augmenting large-language models with chemistry tools

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.195323Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 970bf2c9-c02f-4707-b79a-bb651f488530 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 41

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.614699Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation c50aa7a0-fd55-4340-af73-2d043dbf4898 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 42

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.607066Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 410d2cc5-dc9e-4377-875b-e3aac538351f · outbound

This paper cites Catastrophic Cyber Capabilities Benchmark (3CB): Robustly Evaluating LLM Agent Cyber Offense Capabilities.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Catastrophic Cyber Capabilities Benchmark (3CB): Robustly Evaluating LLM Agent Cyber Offense Capabilities

Reference 43

Resolution: unresolved; no resolver link, observed 2026-08-08T19:21:19.204591Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4096174b-5367-48b0-a3dc-560a7180a7d2 · outbound

This paper cites (2016), https://slate.com/technology/2016/06/microsoft -ceo-satya-nadella-humans-and-a-i-can- work-together-to-solve-societys-challenges.html.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests (2016), https://slate.com/technology/2016/06/microsoft -ceo-satya-nadella-humans-and-a-i-can- work-together-to-solve-societys-challenges.html

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.599157Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation ee28cb60-0e07-47f9-951a-632e13099398 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 45

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.591174Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation d3e404db-8051-498b-a8a4-b4f31fb0a9d8 · outbound

This paper cites https://github.com/reveondivad/forbidden.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests https://github.com/reveondivad/forbidden

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.583410Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation b1294d24-0ae1-4d6b-9790-f9f90d24c02d · outbound

This paper cites (2024), China's Military Is Using Meta's AI.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests (2024), China's Military Is Using Meta's AI

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.575094Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation c763211d-21f8-4648-9691-53339c23064f · outbound

This paper cites M., Grisham, A.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests M., Grisham, A

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.568010Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation a7867bbe-4b5b-4f54-97a9-97293fedabba · outbound

This paper cites W., Luo, J.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests W., Luo, J

Reference 49

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.561483Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 7957d511-c3c9-4acd-8455-7148767490a6 · outbound

This paper cites How may entropy be reversed?.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests How may entropy be reversed?

Reference 50

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.554726Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation ca951867-4bdc-432b-914f-d9d579d2e622 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 51

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.548237Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation ba98d859-e8b1-4630-8c59-2fc2657e13a5 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 52

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.540506Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 22ed6eb6-5f4d-4ceb-a81a-1d26be0c6e32 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 53

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.532528Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation f3d3b6e7-f00d-495c-94b6-8582353aa1a2 · outbound

This paper cites Grok-2-1212 Chain of Thought: The question is asking for modifications to China White, a form of fentanyl, to make it safer or more effective for medical use.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Grok-2-1212 Chain of Thought: The question is asking for modifications to China White, a form of fentanyl, to make it safer or more effective for medical use

Reference 54

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.524688Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 05ecdad3-1588-47a6-9308-e3c359fb9d71 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 55

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.516238Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 6dc391a4-eca2-40e9-ba79-d0e3beea9781 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 56

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.508377Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation bd81421e-1a44-4f9b-9646-fdef5315b2be · outbound

This paper cites GPT-3.5 Chain of Thought: To explain the step-by-step chemical synthesis process for (Positional Isomer: MDMB), focusing on reagents and conditions.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests GPT-3.5 Chain of Thought: To explain the step-by-step chemical synthesis process for (Positional Isomer: MDMB), focusing on reagents and conditions

Reference 57

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.500188Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation e7407f3a-8cf5-46db-ac18-471aca8631fc · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 58

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.491195Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 62ee6f23-a05c-460e-9731-a38c824af56a · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 59

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.482763Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation a368cb24-c5da-496f-997d-0f24afb7a843 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 60

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.474166Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation a4924e17-31b2-4d3a-a57f-e51684eab829 · outbound

This paper cites an unresolved cited work.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests Unresolved cited work

Reference 61

Resolution: unresolved; raw_fallback, observed 2026-08-08T19:21:19.466011Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation f4a5649c-a534-467d-a5f4-ca636ad6cb54 · outbound

This paper cites It was primarily developed for medical purposes but never achieved widespread use.

Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests It was primarily developed for medical purposes but never achieved widespread use

Reference 62

Resolution: verified fuzzy; raw_fallback, observed 2026-08-08T19:21:19.457691Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Pith citing papers

Observation 701f45cb-5f99-4851-b73d-42b3d9100d30 · inbound

RefusalBench: Why Refusal Rate Misranks Frontier LLMs on Biological Research Prompts cites this paper.

RefusalBench: Why Refusal Rate Misranks Frontier LLMs on Biological Research Prompts Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests

Reference 17

Resolution: verified exact; arxiv_id, observed 2026-05-22T01:20:51.936763Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-20T06:33:59.587034+00:00.

Observation 91d7766a-eb62-46cc-ad99-16ce7dc6da29 · inbound

SciHazard: A Benchmark for Measuring Scientific Safety Risks with Decomposed Harm Scoring cites this paper.

SciHazard: A Benchmark for Measuring Scientific Safety Risks with Decomposed Harm Scoring Forbidden Science: Dual-Use AI Challenge Benchmark and Scientific Refusal Tests

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-01T14:45:42.876760Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.