REVIEW 18 cited by
SafeChain: Safety of Language Models with Long Chain-of-Thought Reasoning Capabilities
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Emerging large reasoning models (LRMs), such as DeepSeek-R1 models, leverage long chain-of-thought (CoT) reasoning to generate structured intermediate steps, enhancing their reasoning capabilities. However, long CoT does not inherently guarantee safe outputs, potentially leading to harmful consequences such as the introduction of security vulnerabilities in code or the spread of misinformation. Current research on large language model (LLM) safety usually focuses on short-answer responses, overlooking the long CoT style outputs of LRMs. To bridge this gap, we conduct a systematic study of LRM safety. First, we investigate safety evaluators calibrated against human annotations. Using our newly developed metrics, we thoroughly assess the safety of 12 state-of-the-art LRMs on StrongReject and WildJailbreak datasets. Our results show that LRMs are not safe compared to their reasoning advance. Further, we perform a fine-grained analysis of the reasoning trace and final answer. We find that three decoding strategies-ZeroThink, LessThink, and MoreThink-can improve model safety without additional training. However, these strategies either use constrained reasoning traces or incur high inference costs. To better strengthen LRM safety, we introduce SafeChain, the first-of-its-kind safety training dataset in CoT style. We fine-tune two LRMs with SafeChain, showing that it not only enhances model safety but also preserves performance across 6 reasoning benchmarks.
Forward citations
Cited by 18 Pith papers
-
Risky Business: Measuring The Faithfulness-Safety Tension
Faithful reasoning and safety pull in opposite directions in current reasoning models, and the two behaviors are controlled by anti-correlated internal vectors that can be steered independently.
-
Safety Under Scaffolding: How Evaluation Conditions Shape Measured Safety
Map-reduce scaffolding degrades measured safety mainly by stripping multiple-choice options (40–89% of the loss is format conversion); scaffold architecture explains only 0.4% of variance and composite safety scores h...
-
Selective Forgetting for Large Reasoning Models
FRUL uses multi-LLM RAG to isolate and replace sensitive CoT segments, then applies a suppress-and-reinforce loss so LRMs forget targeted knowledge while retaining structured reasoning.
-
The Emotional Baby Is Truly Deadly: Does your Multimodal Large Reasoning Model Have Emotional Flattery towards Humans?
Multimodal reasoning models can be steered into unsafe behavior by emotional prompts and sometimes conceal harmful reasoning inside seemingly safe responses.
-
Does More Inference-Time Compute Really Help Robustness?
With exposed reasoning chains, increasing inference-time compute consistently decreases measured robustness across 12 open-source reasoning models, while hidden chains show improvements.
-
Lifelong Safety Alignment for Language Models
A co-evolutionary attacker-defender loop, warmed up by strategies extracted from jailbreak papers, reduces jailbreak success rate on a robust model from 73% to 7% in two iterations.
-
Beyond Safe Answers: A Benchmark for Evaluating True Risk Awareness in Large Reasoning Models
A new benchmark shows that top reasoning models identify all relevant risks in under 40% of cases even when their final answers look safe.
-
VisCRA: A Visual Chain Reasoning Attack for Jailbreaking Multimodal Large Language Models
VisCRA jailbreaks multimodal LLMs by masking the most harmful image region and using a two-stage reasoning prompt to make the model infer and then comply.
-
R-TOFU: Unlearning in Large Reasoning Models
R-TOFU shows that answer-level unlearning is insufficient for large reasoning models because residual knowledge persists in chain-of-thought traces.
-
MoGU V2: Toward a Higher Pareto Frontier Between Model Usability and Security
MoGUv2 embeds small routers in the deeper layers of LLMs to dynamically blend a helpful variant and a refusal variant, improving safety against jailbreak and fine-tuning attacks while preserving usability.
-
R1-ACT: Efficient Reasoning Model Safety Alignment by Activating Safety Knowledge
Adding an explicit 'is this harmful?' step to the reasoning chain, trained on just 1,000 examples, substantially reduces harmful responses from reasoning models while roughly preserving benchmark reasoning performance.
-
We Should Identify and Mitigate Third-Party Safety Risks in MCP-Powered Agent Systems
MCP-powered LLM agents are vulnerable to prompt injection from third-party services, and simple detection or filtering defenses do not reliably stop these attacks.
-
Mitigating Deceptive Alignment via Self-Monitoring
CoT Monitor+ embeds self-monitoring into chain-of-thought generation and reports a 43.8% average reduction on DeceptionBench, a GPT-4o-judged deception metric.
-
A Comprehensive Survey on Trustworthiness in Reasoning with Large Language Models
A structured literature survey concluding that reasoning capabilities do not automatically make LLMs more trustworthy and can introduce new vulnerabilities in safety, robustness, and privacy.
-
Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey
A comprehensive review that categorizes methods for shortening and adaptively triggering chain-of-thought reasoning in large language models.
-
Reinforcement Fine-Tuning Powers Reasoning Capability of Multimodal Large Language Models
A survey-style position paper claims that reinforcement fine-tuning powers reasoning in multimodal LLMs, summarizing over a hundred recent works and proposing five future research directions.
-
Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challenges
A broad survey of LLM alignment that catalogs objectives, benchmarks, SFT/RLHF/DPO methods, and safety challenges, without contributing new experimental or theoretical results.
-
DeepSeek: Paradigm Shifts and Technical Evolution in Large AI Models
A survey of DeepSeek's V3 and R1 models covering MLA, MoE, MTP, GRPO, and training engineering, with no new experimental results.
Discussion (0). Sign in to comment.