$D^2$-Monitor: Dynamic Safety Monitoring for Diffusion LLMs via Hesitation-Aware Routing

• URL: http://arxiv.org/abs/2605.25893v1
• Date: Mon, 25 May 2026 14:22:21 GMT
• Title: $D^2$-Monitor: Dynamic Safety Monitoring for Diffusion LLMs via Hesitation-Aware Routing
• Abstract summary: Safety monitoring for large language models (D-LLMs) remains largely unexplored.<n>We propose $D2$-Monitor, a bi-level safety monitor for D-LLMs.<n>$D2$-Monitor adopts a lightweight probe as an always-on monitor to jointly estimate hesitation and perform base classification.
• Score: 63.49501120848927
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Despite the emergence of diffusion large language models (D-LLMs) as an alternative to autoregressive large language models (AR-LLMs), safety monitoring for D-LLMs remains largely unexplored. Unlike AR-LLMs, D-LLMs generate text through a multi-step denoising process, exposing intermediate hidden representations that may contain safety-relevant information unavailable in standard single-step monitoring setups. Motivated by the suitability of lightweight probes for always-on monitoring, we analyze which trajectory-level signals best indicate when such probes are likely to struggle. We find that the most informative signal is safety hesitation: intermediate hidden states repeatedly falling within a small margin of the probe's decision boundary. The number of such hesitation steps in D-LLM's trajectory predicts probe failure effectively, providing a proxy of sample difficulty. Building on this analysis, we propose $D^2$-Monitor, a bi-level safety monitor for D-LLMs. $D^2$-Monitor adopts a lightweight probe as an always-on monitor to jointly estimate hesitation and perform base classification. When the hesitation level exceeds a threshold, a more expressive but computationally heavier probe is activated. This dynamic routing mechanism allocates monitoring resources efficiently at test time. Evaluated on 3 datasets (WildguardMix, ToxicChat, OpenAI-Moderation) across 4 D-LLMs, $D^2$-Monitor achieves state-of-the-art performance with a compact parameter footprint ($\leq$ 0.85M parameters), and exhibits the best trade-off between effectiveness and efficiency relative to 8 baselines.

Related papers

• DeMTS: Denoising Trajectories as Multivariate Time Series for Hallucination Detection in Diffusion Language Models [59.414664850109155]
  Diffusion large language models (D-LLMs) have emerged as a promising paradigm for text generation.<n>D-LLMs remain vulnerable to hallucinations, where fluent outputs may contain factually incorrect or unsupported content.
  arXiv  Detail & Related papers  (2026-07-24T06:45:43Z)
• Constitutional Black-Box Monitoring for Scheming in LLM Agents [1.4619913143519836]
  We use language models to examine agent behaviors for suspicious actions.<n>We study constitutional black-box monitors that detect scheming using only externally observable inputs and outputs.<n>We find that performance saturates quickly in our setting, with simple prompt sweeps matching the results of more extensive optimization.
  arXiv  Detail & Related papers  (2026-02-28T22:31:32Z)
• Beyond Linear Probes: Dynamic Safety Monitoring for Language Models [67.15793594651609]
  Traditional safety monitors require the same amount of compute for every query.<n>We introduce Truncated Polynomials (TPCs), a natural extension of linear probes for dynamic activation monitoring.<n>Our key insight is that TPCs can be trained and evaluated progressively, term-by-term.
  arXiv  Detail & Related papers  (2025-09-30T13:32:59Z)
• TraceDet: Hallucination Detection from the Decoding Trace of Diffusion Large Language Models [49.83690850047884]
  hallucination problem in D-LLMs remains underexplored, limiting their reliability in real-world applications.<n>Existing hallucination detection methods are designed for AR-LLMs and rely on signals from single-step generation.<n>We propose TraceDet, a novel framework that explicitly leverages the intermediate denoising steps of D-LLMs for hallucination detection.
  arXiv  Detail & Related papers  (2025-09-30T02:01:10Z)
• DetectAnyLLM: Towards Generalizable and Robust Detection of Machine-Generated Text Across Domains and Models [60.713908578319256]
  We propose Direct Discrepancy Learning (DDL) to optimize the detector with task-oriented knowledge.<n>Built upon this, we introduce DetectAnyLLM, a unified detection framework that achieves state-of-the-art MGTD performance.<n>MIRAGE samples human-written texts from 10 corpora across 5 text-domains, which are then re-generated or revised using 17 cutting-edge LLMs.
  arXiv  Detail & Related papers  (2025-09-15T10:59:57Z)
• Reliable Weak-to-Strong Monitoring of LLM Agents [6.922769543581406]
  We stress test monitoring systems for detecting covert misbehavior in autonomous agents.<n>We develop a monitor red teaming (MRT) workflow that incorporates varying levels of agent and monitor situational awareness.<n>We release code, data, and logs to spur further research.
  arXiv  Detail & Related papers  (2025-08-26T22:29:31Z)
• Detecting High-Stakes Interactions with Activation Probes [10.421494494629421]
  This paper examines activation probes for detecting "high-stakes" interactions, where the text indicates that the interaction might lead to significant harm.<n>We evaluate several probe architectures trained on synthetic data, and find them to exhibit robust generalization to diverse, out-of-distribution, real-world data.<n>Our experiments also highlight the potential of building resource-aware hierarchical monitoring systems.
  arXiv  Detail & Related papers  (2025-06-12T15:20:33Z)
• Backdoor Cleaning without External Guidance in MLLM Fine-tuning [76.82121084745785]
  Believe Your Eyes (BYE) is a data filtering framework that leverages attention entropy patterns as self-supervised signals to identify and filter backdoor samples.<n>It achieves near-zero attack success rates while maintaining clean-task performance.
  arXiv  Detail & Related papers  (2025-05-22T17:11:58Z)
• You Only Need Two Detectors to Achieve Multi-Modal 3D Multi-Object Tracking [9.20064374262956]
  The proposed framework can achieve robust tracking by using only a 2D detector and a 3D detector. It is proven more accurate than many of the state-of-the-art TBD-based multi-modal tracking methods.
  arXiv  Detail & Related papers  (2023-04-18T02:45:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.