mlsanityData to 14 Oct 2026

TrustLDM: Benchmarking Trustworthiness in Language Diffusion Models

arXiv 2606.00023 ↗First posted 15 Apr 2026cs.CL cs.AI cs.LGcited by 2

On the map: Diffusion models → Discrete and text diffusion

Abstract (arXiv)

The rapid development of Language Diffusion Models (LDMs) challenges the dominant position of auto-regressive competitors in language processing. However, their flexible, any-order decoding strategies not only enable fast decoding speed but also potentially bring new trustworthiness challenges. To better understand the risks behind their pipelines, we introduce a comprehensive trustworthiness benchmark tailored to LDMs (TrustLDM), evaluating safety, privacy, and fairness across different LDM architectures with multiple categories of static post contexts. Our empirical results show that although LDMs generally exhibit strong trustworthiness with only the user prompts, their alignment behavior degrades noticeably when the malicious post contexts are attached to the masked responses. We further observe that longer contexts do not necessarily induce stronger effects, and both decoding order and generation length affect the evaluation outcomes. Finally, we propose TrustLDM-Auto, an automatic evaluation framework that leverages LDM decoding flexibility to systematically identify vulnerable configurations, revealing substantial trustworthiness weaknesses across all evaluated models and dimensions. Our work may potentially help the community build more trustworthy LDMs. Our code is available at https://github.com/PKU-ML/TrustLDM.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Discrete and text diffusion, one of 14 directions in Diffusion models. Builds on 60 of its 68 references that are in the map. First posted on arXiv 15 Apr 2026.

Builds on

60 of 68 references.

Closest ancestors · 9

Same direction, newest first

  1. 2025Decoding Large Language Diffusion Models with Foreseeing Movement3 Dec
  2. 2025LLaDA-MoE: A Sparse MoE Diffusion Language Model29 Sep
  3. 2025Revolutionizing Reinforcement Learning Framework for Diffusion Large Language Models8 Sep
  4. 2025Dream-Coder 7B: An Open Diffusion Language Model for Code1 Sep
  5. 2025Dream 7B: Diffusion Large Language Models21 Aug
  6. 2025Accelerated Sampling from Masked Diffusion Models via Entropy Bounded Unmasking30 May
  7. 2025Large Language Diffusion Models14 Feb
  8. 2025Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusions10 Feb
  9. 2024Scaling Diffusion Language Models via Adaptation from Autoregressive Models23 Oct

Other references · 51

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  3. 2023GPT-4 Technical Reportcited by 18,396
  4. 2024The Llama 3 Herd of Modelscited by 14,715
  5. 2021Diffusion Models Beat GANs on Image Synthesiscited by 8,255
  6. 2024GPT-4o System Cardcited by 4,588
  7. 2024Scaling Rectified Flow Transformers for High-Resolution Image Synthesiscited by 4,181
  8. 2023Universal and Transferable Adversarial Attacks on Aligned Language Modelscited by 3,104
  9. 2022Elucidating the Design Space of Diffusion-Based Generative Modelscited by 3,045
  10. 2022Video Diffusion Modelscited by 2,305
  11. 2016Equality of Opportunity in Supervised Learningcited by 2,007
  12. 2024Mixtral of Expertscited by 1,823
38 more
  1. 2023Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Modelscited by 1,476
  2. 2024HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusalcited by 1,429
  3. 2023Jailbreaking Black Box Large Language Models in Twenty Queriescited by 1,420
  4. 2024WorldSimBench: Towards Video Generation Models as World Simulatorscited by 881
  5. 2023Tree of Attacks: Jailbreaking Black-Box LLMs Automaticallycited by 672
  6. 2023Structure and Content-Guided Video Synthesis with Diffusion Modelscited by 590
  7. 2023GPTFUZZER: Red Teaming Large Language Models with Auto-Generated Jailbreak Promptscited by 560
  8. 2024JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Modelscited by 536
  9. 2023DecodingTrust: A Comprehensive Assessment of Trustworthiness in GPT Modelscited by 518
  10. 2022Diffusion Models for Adversarial Purificationcited by 476
  11. 2024Safety Alignment Should Be Made More Than Just a Few Tokens Deepcited by 472
  12. 2023The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"cited by 407
  13. 2023I2VGen-XL: High-Quality Image-to-Video Synthesis via Cascaded Diffusion Modelscited by 350
  14. 2024TrustLLM: Trustworthiness in Large Language Modelscited by 285
  15. 2025LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Modelscited by 285
  16. 2025Mercury: Ultra-Fast Language Models Based on Diffusioncited by 235
  17. 2023Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theorycited by 230
  18. 2024JailBreakV: A Benchmark for Assessing the Robustness of MultiModal Large Language Models against Jailbreak Attackscited by 230
  19. 2025d1: Scaling Reasoning in Diffusion Large Language Models via Reinforcement Learningcited by 186
  20. 2025LLaDA-V: Large Language Diffusion Models with Visual Instruction Tuningcited by 148
  21. 2024Diffusion Language Models Are Versatile Protein Learnerscited by 98
  22. 2025LaViDa: A Large Diffusion Language Model for Multimodal Understandingcited by 84
  23. 2025Wide-In, Narrow-Out: Revokable Decoding for Efficient and Effective DLLMscited by 46
  24. 2024LLM-PBE: Assessing Data Privacy in Large Language Modelscited by 41
  25. 2023PrivLM-Bench: A Multi-level Privacy Evaluation Benchmark for Language Modelscited by 37
  26. 2025The Devil behind the mask: An emergent safety vulnerability of Diffusion LLMscited by 34
  27. 2023How Trustworthy are Open-Source LLMs? An Assessment under Malicious Demonstrations Shows their Vulnerabilitiescited by 33
  28. 2024Text-Guided Molecule Generation with Diffusion Language Modelcited by 31
  29. 2023BerDiff: Conditional Bernoulli Diffusion Model for Medical Image Segmentationcited by 27
  30. 2025Jailbreaking Large Language Diffusion Models: Revealing Hidden Safety Flaws in Diffusion-Based Text Generationcited by 26
  31. 2024Generated Data with Fake Privacy: Hidden Dangers of Fine-tuning Large Language Models on Generated Datacited by 22
  32. 2025Are Smarter LLMs Safer? Exploring Safety-Reasoning Trade-offs in Prompting and Fine-Tuningcited by 20
  33. 2024CEB: Compositional Evaluation Benchmark for Fairness in Large Language Modelscited by 18
  34. 2025Adjacent Words, Divergent Intents: Jailbreaking Large Language Models via Task Concurrencycited by 17
  35. 2023Score dynamics: scaling molecular dynamics with picoseconds timestep via conditional diffusion modelcited by 16
  36. 2025JADES: A Universal Framework for Jailbreak Assessment via Decompositional Scoringcited by 14
  37. 2024Text-guided multi-property molecular optimization with a diffusion language modelcited by 11
  38. 2025FLEX: A Benchmark for Evaluating Robustness of Fairness in Large Language Modelscited by 8

50 most cited of 51 All 51 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026HarmfulSkillBench: How Do Harmful Skills Weaponize Your Agents?cited by 11
  2. 2026Noise Out, Bias In: Targeted Bias Injection in Diffusion Language Models via Closed-Loop Activation Steeringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Towards Tracing Trustworthiness Dynamics: Revisiting Pre-training Period of Large Language Models
  2. 2026Benchmarking Trustworthiness of SLMs: Pre-trained vs. Compressed
  3. 2024Decoding Compressed Trust: Scrutinizing the Trustworthiness of Efficient LLMs Under Compression
  4. 2025Unveiling Trust in Multimodal Large Language Models: Evaluation, Analysis, and Mitigation
  5. 2025Toward Safer Diffusion Language Models: Discovery and Mitigation of Priming Vulnerability
  6. 2024MultiTrust: A Comprehensive Benchmark Towards Trustworthy Multimodal Large Language Models
  7. 2026Membership Inference Attacks Against Fine-tuned Diffusion Language Models
  8. 2026BadDLM: Backdooring Diffusion Language Models with Diverse Targets