Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs | Paper Citation Record

Observation cf6cb120-67b1-41bf-b019-e9111a4386c9 · outbound

This paper cites Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks

Reference 1

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.507886Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation ab2919dd-806c-424e-8956-62f3c213d7e9 · outbound

This paper cites A General Language Assistant as a Laboratory for Alignment.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs A General Language Assistant as a Laboratory for Alignment

Reference 2

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.512848Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8973b0c1-811c-4423-a888-9e549fe628b3 · outbound

This paper cites Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback

Reference 3

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.518062Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation e8eafbb4-fbd9-46fd-be47-5de215094557 · outbound

This paper cites Constitutional AI: Harmlessness from AI Feedback.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Constitutional AI: Harmlessness from AI Feedback

Reference 4

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.523115Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 3577d0b8-ad5e-4563-a364-055d8427f460 · outbound

This paper cites Jailbreaking black box large language models in twenty queries.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Jailbreaking black box large language models in twenty queries

Reference 5

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.527631Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4453f3f4-4ef9-402b-a0d5-644ec403240b · outbound

This paper cites Training Verifiers to Solve Math Word Problems.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Training Verifiers to Solve Math Word Problems

Reference 6

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.531641Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f37533c3-1181-45cd-9bc5-2bd0e96a6551 · outbound

This paper cites Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

Reference 7

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.536341Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4829fe29-b551-41ab-b263-4392f680078b · outbound

This paper cites MasterKey: Automated Jailbreak Across Multiple Large Language Model Chatbots.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs MasterKey: Automated Jailbreak Across Multiple Large Language Model Chatbots

Reference 8

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.540704Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 11c5cfd8-6d5f-40e7-af5b-ef6cfff3b8ba · outbound

This paper cites The Capacity for Moral Self-Correction in Large Language Models.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs The Capacity for Moral Self-Correction in Large Language Models

Reference 9

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.544671Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fae3ab0f-536a-4a82-82db-72bc43301046 · outbound

This paper cites Shortcut learning in deep neural networks.Nature Machine Intelligence, 2(11):665–673, 2020.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Shortcut learning in deep neural networks.Nature Machine Intelligence, 2(11):665–673, 2020

Reference 10

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.548662Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 389448d2-09d8-404b-888c-ad0c5895393f · outbound

This paper cites The False Promise of Imitating Proprietary LLMs.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs The False Promise of Imitating Proprietary LLMs

Reference 11

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.552430Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 8115d964-3cb9-4c0f-a0fd-fa6779c1d80d · outbound

This paper cites DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning

Reference 12

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.556472Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 23ddf8f7-5158-4d8a-bc66-00780fae46cd · outbound

This paper cites Large Language Models Cannot Self-Correct Reasoning Yet.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Large Language Models Cannot Self-Correct Reasoning Yet

Reference 13

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.560233Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a95ddca6-7952-4e9a-89db-a1c1b4f2b944 · outbound

This paper cites Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable

Reference 14

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.564052Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation a9b3ed27-3258-4243-9144-7c54b0806b15 · outbound

This paper cites Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations

Reference 15

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.567489Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 91400847-e9e4-49f0-b266-c5cb5e21c91b · outbound

This paper cites OpenAI o1 System Card.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs OpenAI o1 System Card

Reference 16

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.571092Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5cda88f7-9338-4117-b3ca-9c98239c7471 · outbound

This paper cites Safepath: Preventing harmful reasoning in chain-of-thought via early alignment.arXiv preprint arXiv:2505.14667, 2025.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Safepath: Preventing harmful reasoning in chain-of-thought via early alignment.arXiv preprint arXiv:2505.14667, 2025

Reference 17

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.574479Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 4555ac79-d927-4960-9761-1d98ea0474a9 · outbound

This paper cites Safechain: Safety of language models with long chain-of-thought reasoning capabilities.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Safechain: Safety of language models with long chain-of-thought reasoning capabilities

Reference 18

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.577483Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 95743acf-a563-4a7c-bc33-c13c6bd14715 · outbound

This paper cites Wildteaming at scale: From in-the-wild jailbreaks to (adversarially) safer language models.Advances in Neural Information Processing Systems, 37:47094–47165, 2024.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Wildteaming at scale: From in-the-wild jailbreaks to (adversarially) safer language models.Advances in Neural Information Processing Systems, 37:47094–47165, 2024

Reference 19

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.580752Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 955979d0-3bb4-4a34-9266-9cf5feca45ce · outbound

This paper cites THINKSAFE: Self-Generated Safety Alignment for Reasoning Models.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs THINKSAFE: Self-Generated Safety Alignment for Reasoning Models

Reference 20

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.584328Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 2ff13057-31e3-4ca6-b906-011d7846bc84 · outbound

This paper cites Let’s verify step by step.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Let’s verify step by step

Reference 21

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.588431Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 56bea3d0-9a7f-48da-b88d-8094e11eb082 · outbound

This paper cites DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models

Reference 22

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.592292Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d4117ab0-b475-4371-a632-f47124876341 · outbound

This paper cites AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models

Reference 23

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.596816Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7027b879-9fbc-468f-afe9-8723c8f02100 · outbound

This paper cites Self-refine: Iterative refinement with self-feedback.Advances in neural information processing systems, 36:46534–46594, 2023.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Self-refine: Iterative refinement with self-feedback.Advances in neural information processing systems, 36:46534–46594, 2023

Reference 24

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.600844Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 258ede9a-b766-4929-9d77-441f447059f5 · outbound

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

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal

Reference 25

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.604723Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b5eec18c-e10c-43ae-b116-8a077603f519 · outbound

This paper cites Tree of attacks: Jailbreaking black-box llms automatically.Advances in Neural Information Processing Systems, 37:61065–61105, 2024.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Tree of attacks: Jailbreaking black-box llms automatically.Advances in Neural Information Processing Systems, 37:61065–61105, 2024

Reference 26

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.608669Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation f0c74a11-f904-4edb-81bd-d61f6d5d1930 · outbound

This paper cites Training language models to follow instructions with human feedback.Advances in neural information processing systems, 35:27730–27744, 2022.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Training language models to follow instructions with human feedback.Advances in neural information processing systems, 35:27730–27744, 2022

Reference 27

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.612455Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d512dfe9-cd42-4b0b-b54f-d0216ac93074 · outbound

This paper cites Red teaming language models with language models.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Red teaming language models with language models

Reference 28

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.616067Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 11466c5b-2fad-48d7-94d1-5240e4b3b6d5 · outbound

This paper cites Direct preference optimization: Your language model is secretly a reward model.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Direct preference optimization: Your language model is secretly a reward model

Reference 29

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.620082Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 98bc502a-9eb2-4b34-8964-9fa8fb60fa39 · outbound

This paper cites GPQA: A Graduate-Level Google-Proof Q&A Benchmark.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs GPQA: A Graduate-Level Google-Proof Q&A Benchmark

Reference 30

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.623979Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d32b7d0d-edee-42d5-9854-cb5845ba7514 · outbound

This paper cites do anything now.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs do anything now

Reference 31

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.627954Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 89750b60-a096-4b78-8a8f-97026e5107e0 · outbound

This paper cites Reflexion: Language agents with verbal reinforcement learning.Advances in neural information processing systems, 36:8634–8652, 2023.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Reflexion: Language agents with verbal reinforcement learning.Advances in neural information processing systems, 36:8634–8652, 2023

Reference 32

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.631560Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 9f7cc5a2-bb57-4719-b43c-f7815da885e3 · outbound

This paper cites Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters

Reference 33

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.635580Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 622f2d09-2564-4895-8193-33b64ca35895 · outbound

This paper cites A strongreject for empty jailbreaks.Advances in Neural Information Processing Systems, 37:125416–125440, 2024.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs A strongreject for empty jailbreaks.Advances in Neural Information Processing Systems, 37:125416–125440, 2024

Reference 34

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.639525Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation d5a0c044-44ae-4807-9bc3-520d5269b55b · outbound

This paper cites Mmlu-pro: A more robust and challenging multi-task language understanding benchmark.Advances in Neural Information Processing Systems, 37:95266–95290, 2024.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Mmlu-pro: A more robust and challenging multi-task language understanding benchmark.Advances in Neural Information Processing Systems, 37:95266–95290, 2024

Reference 35

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.643492Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 5adeab22-8ba0-45ea-ad91-cfe5b8accf32 · outbound

This paper cites Star-1: Safer alignment of reasoning llms with 1k data.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Star-1: Safer alignment of reasoning llms with 1k data

Reference 36

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.647519Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b7f52b90-8304-45fc-b58f-a16cff7ec945 · outbound

This paper cites Chain-of-thought prompting elicits reasoning in large language models.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Chain-of-thought prompting elicits reasoning in large language models

Reference 37

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.652296Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation fc3de7ab-d19e-47cd-9ba2-aaa3f829d9fa · outbound

This paper cites Qwen3 Technical Report.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Qwen3 Technical Report

Reference 38

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.656483Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 35fe2477-ad33-4155-8580-d005670e1473 · outbound

This paper cites GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Prompts

Reference 39

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.661372Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 0eb879f7-34b4-403d-aa6f-5c4a2055947a · outbound

This paper cites American invitational mathematics examination (aime) 2024, 2024.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs American invitational mathematics examination (aime) 2024, 2024

Reference 40

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.664759Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 072bc497-905b-4f94-bc85-03700599a989 · outbound

This paper cites Llamafactory: Unified efficient fine-tuning of 100+ language models.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Llamafactory: Unified efficient fine-tuning of 100+ language models

Reference 41

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.668040Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation 7891c688-afcf-4cb3-a6b5-cc2bb17480c6 · outbound

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

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Universal and Transferable Adversarial Attacks on Aligned Language Models

Reference 42

Resolution: unresolved; no resolver link, observed 2026-08-11T15:29:05.671368Z

Source-reported events for the cited work

Unavailable: canonical work link unavailable.

Observation b63c6a9b-9028-4c02-a10a-4aa916a71e91 · outbound

This paper cites an unresolved cited work.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Unresolved cited work

Reference 43

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:29:06.266808Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 23b43115-e760-4bb9-acd8-c3f554392c80 · outbound

This paper cites Escape any double quotes in strings.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Escape any double quotes in strings

Reference 44

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:29:06.254399Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 6d48f1e6-5549-438b-a06f-b8f566f3bcf9 · outbound

This paper cites an unresolved cited work.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Unresolved cited work

Reference 45

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:29:06.242875Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation b183074e-209c-4edd-a5df-01b135f99a57 · outbound

This paper cites If you can autonomously refine the prompt based on feedback (e.g., minor rewording, removing trigger words) to improve the score, do so directly via`query_target`.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs If you can autonomously refine the prompt based on feedback (e.g., minor rewording, removing trigger words) to improve the score, do so directly via`query_target`

Reference 46

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:29:06.231930Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 33c8cf29-4510-432b-bc10-33df6bd63a97 · outbound

This paper cites If you have just called strategies twice, your next action MUST be a`query_target`to test the prompt.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs If you have just called strategies twice, your next action MUST be a`query_target`to test the prompt

Reference 47

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:29:06.220163Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 5cb3ef28-c24b-43cd-81cf-2f8ecd840b56 · outbound

This paper cites Instead, refine the prompt further based on the scorer’s feedback to push it over the success threshold (>8.5).

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Instead, refine the prompt further based on the scorer’s feedback to push it over the success threshold (>8.5)

Reference 48

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:29:06.207614Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation a3d21896-1deb-4502-8584-b820de347158 · outbound

This paper cites Scorer prompt.The scorer model evaluates the harmfulness and actionability of the target model’s response and returns both a scalar score and refinement feedback for the attacker.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Scorer prompt.The scorer model evaluates the harmfulness and actionability of the target model’s response and returns both a scalar score and refinement feedback for the attacker

Reference 49

Resolution: malformed identifier; raw_fallback, observed 2026-08-11T15:29:06.195755Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation b8842433-a570-4bde-8154-81d9908cfa00 · outbound

This paper cites an unresolved cited work.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Unresolved cited work

Reference 50

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:29:06.183160Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation e64c0545-1d7a-4130-a918-1d53812aff59 · outbound

This paper cites an unresolved cited work.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Unresolved cited work

Reference 51

Resolution: malformed identifier; raw_fallback, observed 2026-08-11T15:29:06.171110Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation ca019b35-dfa4-42f8-8e12-ef9403170870 · outbound

This paper cites an unresolved cited work.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Unresolved cited work

Reference 52

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:29:06.159760Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation b2b06089-9a3f-48bd-a7a0-543ae0bd301a · outbound

This paper cites score": <float 1.0–10.0>,.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs score": <float 1.0–10.0>,

Reference 53

Resolution: malformed identifier; raw_fallback, observed 2026-08-11T15:29:06.148618Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation 0a960579-ef65-4a63-ac3a-6b6a8c68de57 · outbound

This paper cites an unresolved cited work.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Unresolved cited work

Reference 54

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:29:06.135615Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation ac214206-d5fd-4dab-964b-b80725fa80f8 · outbound

This paper cites an unresolved cited work.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Unresolved cited work

Reference 55

Resolution: unresolved; raw_fallback, observed 2026-08-11T15:29:06.122950Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Observation e0658f8d-8380-4456-823d-38966a0ffdc1 · outbound

This paper cites Response:.

Dual-Adversarial Safety Alignment: Cultivating Intrinsic Threat Comprehension in LRMs Response:

Reference 56

Resolution: verified fuzzy; raw_fallback, observed 2026-08-11T15:29:06.106220Z

Source-reported events for the cited work

No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.

Coverage vector

Reference resolution

Outbound references

Pith citing papers