A survey and comparative analysis of NLP-based Automatic Deception Detection focusing on the legal domain and the evolution from feature-based machine learning to Large Language Model (LLM) approaches are presented, showing strong domain sensitivity.
Abstract
Deception detection has critical implications for legal proceedings, law enforcement, and online security. Although human judgment is limited in accuracy and scalability, Natural Language Processing (NLP) offers a data-driven alternative. We present a survey and comparative analysis of NLP-based Automatic Deception Detection (ADD) focusing on the legal domain, reviewing the evolution from feature-based machine learning to Large Language Model (LLM) approaches. We conduct a unified empirical evaluation across seven datasets (two legal, five general-domain), comparing six fine-tuned transformer models and seven LLMs under four prompting strategies. The results show strong domain sensitivity, with fine-tuned models excelling in data-rich general domains and few-shot LLMs remaining competitive in low-resource legal settings. Chain-of-Thought prompting often underperforms direct classification. These findings highlight the need for domain adaptation and interpretable systems in high-stakes legal contexts.
Large language models (LLMs) have demonstrated strong performance across a wide range of legal tasks. In legal practice, time is a critical concept that governs the validity of statutes, the progression of legal cases, and the enforcement of procedural deadlines. However, legal temporal capabilities remain underexplored in existing legal AI benchmarks. To address this gap, we propose LexKairos, a comprehensive benchmark for evaluating the temporal capabilities of LLMs in the Chinese legal context across three dimensions: statutory temporal knowledge, case temporal modeling, and statute-case temporal reasoning. LexKairos comprises nine sub-tasks drawn from real-world Chinese judicial cases and statutes. We conduct systematic evaluations of eight LLMs under multiple inference settings, including vanilla, Chain-of-Thought (CoT), and thinking modes. Our results show that Gemini-3-Flash achieves the strongest overall performance, yet even the best-performing model exhibits notable limitations on tasks demanding precise time-sensitive statutory metadata recall or complex reasoning in time limits, indicating that legal temporal knowledge and reasoning remain open challenges for current LLMs. Data and code are available at https://github.com/thunlp/LexKairos.
Chenyang Li, Zejia Feng, Yuqi Huang et al.· 0 citations
This study evaluates the applicability of generative artificial intelligence (AI), specifically Large Language Models (LLMs), in legal assistance tasks using the Artificial Intelligence for Legal Assistance challenge dataset, which includes 197 statutes, 2914 judicial cases, and 50 legal queries. The research compared conversational LLM tools with semantic retrieval systems based on dense embeddings. Results show that general-purpose tools (ChatGPT, NotebookLM) achieved poor discrimination, while embedding-based methods significantly improved accuracy: the OpenAI text-embedding-ada-002 model reached 46.43% retrieval accuracy, and a fine-tuned all-mpnet-base-v2 model improved from 18.9% to 31.87% (a 68.6% relative gain). These findings demonstrate that combining LLMs with semantic retrieval techniques enhances precision and scalability in legal information systems, offering a viable roadmap for developing domain-specific, efficient, and sustainable AI legal assistants.
Carlos A. Martín, J. Torres, R. M. Aguilar et al.· Logic Journal of the IGPL· 0 citations
Large Language Models (LLMs) are increasingly used to automate high-stakes screening decisions, yet they can be manipulated by adversarial instructions hidden in the documents they evaluate. This paper introduces a benchmark for this vulnerability in LLM-based resume screening: 463 job-candidate pairs drawn from a 14-domain corpus, with the evaluated sample covering 13 domains, attacked through a taxonomy of four attack types and four injection positions (16 attack configurations). Across 12 model configurations covering open-weight and proprietary models, some attack types exceed 80% attack success rate (ASR) when the injected content reaches the model, and attacks upgrade up to 73.4% of candidates unanimously rejected by human annotators. The hidden-content attacks assume the resume text or HTML reaches the model; an end-to-end parser-layer analysis shows that style-aware sanitization removes most of them before classification, whereas visible-text attacks survive. We evaluate prompt-based defenses on all 12 model configurations and our proposed FIDS (Foreign Instruction Detection through Separation), a fine-tuning defense, on Qwen3-8B, the one base model we could fine-tune. On Qwen3-8B, in paired configuration-level comparisons against a common no-defense baseline, prompt-based defense reduces ASR by 10.1 percentage points (pp; 95% bootstrap CI [6.3, 14.3]), FIDS by 15.4 pp (95% CI [8.8, 23.4]), and their combination by 26.3 pp (95% CI [18.2, 35.0]), at the cost of also downgrading candidates the undefended model had accepted (a proxy for false rejections), by 12.5, 10.4, and 19.4 pp respectively. No defense eliminates the attacks, training-time and prompt-only defenses have comparable utility costs, and whether these trade-offs carry over to the proprietary models we cannot fine-tune is left to future work.
Hong-Lin Mu, Jinghao Liu, Kaiyang Wan et al.· International Journal of Mac...· 3 citations· ⚡1
This work proposes an explanation-aware safety framework that augments binary harmfulness detection with structured, human-interpretable explanations capturing severity, strategies, trigger spans, ratio-nales, and derived safety factors, and introduces a human–LLM hybrid annotation and canonicaliza-tion pipeline.
Sunghee Dong, Sungwon Yi, Kangmin Bae et al.· 0 citations
DeBERTa-Sentinel is introduced, a responsible AI-generated text detection framework leveraging DeBERTa-v3's disentangled attention to capture subtle structural irregularities in synthetic content and promotes trustworthy, ethical, and human-centric AI systems.
Detecting whether a jailbreak attack against a large language model (LLM) has succeeded is a challenging problem in AI security. Existing approaches rely on brittle string matching, expensive LLM-based judges, or task-specific classifiers that often fail to generalize. This paper investigates whether semantic coherence is sufficient to evaluate jailbreak success. In this work, we propose PRISM: Prompt-Response coherence for Identifying jailbreaks via Scoring Method, a lightweight method that models jailbreak detection as a coherence problem between the harmful prompt and the model response. The approach combines multiple semantic similarity signals within an ensemble of classical and simple machine learning models. Training in 3,480 pairs of prompt-responses from AdvBench, HarmBench and Alpaca, PRISM achieves F1 = 0.91 and AUC-ROC = 0.98 in the internal test set. In a held-out validation set of 694 labeled samples never seen during training, it outperforms six baseline attack success estimators, achieving the highest accuracy (0.746) and AUC-ROC (0.840), while being up to $\mathbf{1 6} \times$ faster than LLM-based methods. These results demonstrate that semantic coherence provides a robust and efficient signal for jailbreak detection, significantly reducing the implementation complexity of state-of-the-art approaches.
Costabile Di Gregorio, Lama Sleem, Jérôme François et al.· Annual International Compute...· 0 citations
