The Regularizing Power of Language-Training Deepfake Detectors

• URL: http://arxiv.org/abs/2605.31192v1
• Date: Fri, 29 May 2026 12:01:17 GMT
• Title: The Regularizing Power of Language-Training Deepfake Detectors
• Abstract summary: Deepfake detection generally suffers from overfitting to low-level domain-specific artifacts.<n>We use a dual-encoder architecture, pairing a frozen specialist detector with a LoRA-tuned MLLM encoder.<n>We employ a reinforcement learning stage that encourages the model to generate descriptive reasoning before classifying.
• Score: 57.529452351298495
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recently, thanks to the advent of Multimodal-LLMs, deepfake detectors are striving not only to be generalizable but also interpretable. We propose that these two challenges can effectively be tackled jointly, since describable artifacts typically generalize better, opening the possibility to use language as a regularization mechanism. Since deepfake detection generally suffers from overfitting to low-level domain-specific artifacts, our intuition is that an LLM that has been pretrained on language would prefer high-level artifacts that can be described better. This way, we can use high-level features where possible, while training the model to use low-level features where necessary. We utilize a dual-encoder architecture, pairing a frozen specialist detector with a LoRA-tuned MLLM encoder, and a two-stage training curriculum: first, a binary alignment phase demonstrates that the intrinsic capability of MLLMs can effectively combine features to mitigate overfitting to dataset-specific artifacts. To further bolster generalization and achieve interpretability, we employ a reinforcement learning stage that encourages the model to generate descriptive reasoning before classifying, using only binary labels. By rewarding this "explain-then-classify" behavior, we explicitly incentivize the model to prioritize high-level, robust features. Crucially, this process yields both interpretable descriptions and a further boost in cross-dataset performance, even when reasoning chains are omitted at inference. Extensive experiments on benchmark datasets validate our approach, outperforming state-of-the-art methods by a large margin.

Related papers

• Deep Residual Injection for Full-Spectrum Forensic Signal Perception in Multimodal Large Language Models [51.099179436350376]
  Multimodal large language models (MLLMs) have been increasingly adopted in forensics for their robust semantic understanding.<n>We propose Deep Visual Residual MLLM (Deep-VRM) to preserve early semantic processing while injecting artifact-specific visual signals as a residual path into an intermediate layer.<n>This enables later layers to jointly model semantic reasoning and signal-level forensic cues, and surprisingly, the model learns to adaptively leverage different levels of forensic signals depending on the input.
  arXiv  Detail & Related papers  (2026-06-14T16:05:45Z)
• Reasoning-Aware AIGC Detection via Alignment and Reinforcement [55.09684020007737]
  REVEAL is a framework that generates interpretable reasoning chains before classification.<n>It achieves state-of-the-art performance across multiple benchmarks.
  arXiv  Detail & Related papers  (2026-04-21T07:29:55Z)
• ICLAD: In-Context Learning with Comparison-Guidance for Audio Deepfake Detection [7.887913982488066]
  We introduce a novel textbfIn-textbfContext textbfLearning paradigm with comparison-guidance for textbfAudio textbfDeepfake detection.<n>The framework enables the use of audio language models (ALMs) for training-free generalization to unseen deepfakes.
  arXiv  Detail & Related papers  (2026-04-17T23:44:33Z)
• Revisiting Unknowns: Towards Effective and Efficient Open-Set Active Learning [32.530220943884665]
  Open-set active learning (OSAL) aims to identify informative samples for annotation when unlabeled data may contain previously unseen classes.<n>We propose E$2$OAL (Effective and Efficient Open-set Active Learning), a unified and detector-free framework.<n>E$2$OAL fully exploits labeled unknowns for both stronger supervision and more reliable querying.
  arXiv  Detail & Related papers  (2026-03-09T02:35:36Z)
• Deepfake Detection that Generalizes Across Benchmarks [48.85953407706351]
  The generalization of deepfake detectors to unseen manipulation techniques remains a challenge for practical deployment.<n>This work demonstrates that robust generalization is achievable through a parameter-efficient adaptation of one of the foundational pre-trained vision encoders.<n>The proposed method achieves state-of-the-art performance, outperforming more complex, recent approaches in average cross-dataset AUROC.
  arXiv  Detail & Related papers  (2025-08-08T12:03:56Z)
• X2-DFD: A framework for eXplainable and eXtendable Deepfake Detection [55.77552681618732]
  X2-DFD is an eXplainable and eXtendable framework based on multimodal large-language models (MLLMs) for deepfake detection.<n>The first stage, Model Feature Assessment, systematically evaluates the detectability of forgery-related features for the MLLM.<n>The second stage, Explainable dataset Construction, consists of two key modules: Strong Feature Strengthening and Weak Feature Supplementing.<n>The third stage, Fine-tuning and Inference, involves fine-tuning the MLLM on the constructed dataset and deploying it for final detection and explanation.
  arXiv  Detail & Related papers  (2024-10-08T15:28:33Z)
• Conditional Prototype Rectification Prompt Learning [32.533844163120875]
  We propose a Prototype Rectification Prompt Learning (CPR) method to correct the bias of base examples and augment limited data in an effective way. CPR achieves state-of-the-art performance on both few-shot classification and base-to-new generalization tasks.
  arXiv  Detail & Related papers  (2024-04-15T15:43:52Z)
• Bidirectional Trained Tree-Structured Decoder for Handwritten Mathematical Expression Recognition [51.66383337087724]
  The Handwritten Mathematical Expression Recognition (HMER) task is a critical branch in the field of OCR. Recent studies have demonstrated that incorporating bidirectional context information significantly improves the performance of HMER models. We propose the Mirror-Flipped Symbol Layout Tree (MF-SLT) and Bidirectional Asynchronous Training (BAT) structure.
  arXiv  Detail & Related papers  (2023-12-31T09:24:21Z)
• Slot Induction via Pre-trained Language Model Probing and Multi-level Contrastive Learning [62.839109775887025]
  Slot Induction (SI) task whose objective is to induce slot boundaries without explicit knowledge of token-level slot annotations. We propose leveraging Unsupervised Pre-trained Language Model (PLM) Probing and Contrastive Learning mechanism to exploit unsupervised semantic knowledge extracted from PLM. Our approach is shown to be effective in SI task and capable of bridging the gaps with token-level supervised models on two NLU benchmark datasets.
  arXiv  Detail & Related papers  (2023-08-09T05:08:57Z)

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.