Beyond Token Probes: Hallucination Detection via Activation Tensors with ACT-ViT

• URL: http://arxiv.org/abs/2510.00296v1
• Date: Tue, 30 Sep 2025 21:37:43 GMT
• Title: Beyond Token Probes: Hallucination Detection via Activation Tensors with ACT-ViT
• Authors: Guy Bar-Shalom, Fabrizio Frasca, Yaniv Galron, Yftah Ziser, Haggai Maron,
• Abstract summary: We build on the natural sequential structure of activation data in both axes.<n>We advocate treating full activation tensors akin to images.<n>We show that our architecture benefits substantially from multi-LLM training.
• Score: 30.106500467531873
• License: http://creativecommons.org/licenses/by-nc-nd/4.0/
• Abstract: Detecting hallucinations in Large Language Model-generated text is crucial for their safe deployment. While probing classifiers show promise, they operate on isolated layer-token pairs and are LLM-specific, limiting their effectiveness and hindering cross-LLM applications. In this paper, we introduce a novel approach to address these shortcomings. We build on the natural sequential structure of activation data in both axes (layers $\times$ tokens) and advocate treating full activation tensors akin to images. We design ACT-ViT, a Vision Transformer-inspired model that can be effectively and efficiently applied to activation tensors and supports training on data from multiple LLMs simultaneously. Through comprehensive experiments encompassing diverse LLMs and datasets, we demonstrate that ACT-ViT consistently outperforms traditional probing techniques while remaining extremely efficient for deployment. In particular, we show that our architecture benefits substantially from multi-LLM training, achieves strong zero-shot performance on unseen datasets, and can be transferred effectively to new LLMs through fine-tuning. Full code is available at https://github.com/BarSGuy/ACT-ViT.

Related papers

• A Comprehensive Study on Visual Token Redundancy for Discrete Diffusion-based Multimodal Large Language Models [85.30893355216486]
  We study how visual token redundancy evolves with different dMLLM architectures and tasks.<n>Our study reveals that visual redundancy emerges only in from-scratch dMLLMs while handling long-answer tasks.<n>Layer-skipping is promising for accelerating AR-to-diffusion dMLLMs, whereas progressive or late-step pruning is more effective for from-scratch dMLLMs.
  arXiv  Detail & Related papers  (2025-11-19T04:13:36Z)
• Beyond CNNs: Efficient Fine-Tuning of Multi-Modal LLMs for Object Detection on Low-Data Regimes [0.0]
  We compare fine-tuned traditional CNNs, zero-shot pre-trained multi-modal LLMs, and fine-tuned multi-modal LLMs on the challenging task of artificial text overlay detection in images.<n>A key contribution of our study is demonstrating that LLMs can be effectively fine-tuned on very limited data to achieve up to 36% accuracy improvement.<n>Our work contributes to the broader effort of bridging vision and language, offering novel insights into efficient cross-modal learning strategies.
  arXiv  Detail & Related papers  (2025-10-03T18:53:18Z)
• Enhancing Cross-task Transfer of Large Language Models via Activation Steering [75.41750053623298]
  Cross-task in-context learning offers a direct solution for transferring knowledge across tasks.<n>We investigate whether cross-task transfer can be achieved via latent space steering without parameter updates or input expansion.<n>We propose a novel Cross-task Activation Steering Transfer framework that enables effective transfer by manipulating the model's internal activation states.
  arXiv  Detail & Related papers  (2025-07-17T15:47:22Z)
• SAILViT: Towards Robust and Generalizable Visual Backbones for MLLMs via Gradual Feature Refinement [11.815369617016174]
  Vision Transformers (ViTs) are essential as foundation backbones in establishing the visual comprehension capabilities of Multimodal Large Language Models (MLLMs)<n>This paper proposes a gradual feature learning-enhanced ViT for facilitating MLLMs to break through performance bottlenecks in complex multimodal interactions.
  arXiv  Detail & Related papers  (2025-07-02T12:17:23Z)
• Mitigating Forgetting in LLM Fine-Tuning via Low-Perplexity Token Learning [65.23593936798662]
  We show that fine-tuning with LLM-generated data improves target task performance and reduces non-target task degradation.<n>This is the first work to provide an empirical explanation based on token perplexity reduction to mitigate catastrophic forgetting in LLMs after fine-tuning.
  arXiv  Detail & Related papers  (2025-01-24T08:18:56Z)
• BANER: Boundary-Aware LLMs for Few-Shot Named Entity Recognition [12.57768435856206]
  We propose an approach called Boundary-Aware LLMs for Few-Shot Named Entity Recognition.<n>We introduce a boundary-aware contrastive learning strategy to enhance the LLM's ability to perceive entity boundaries for generalized entity spans.<n>We utilize LoRAHub to align information from the target domain to the source domain, thereby enhancing adaptive cross-domain classification capabilities.
  arXiv  Detail & Related papers  (2024-12-03T07:51:14Z)
• Accelerating Multimodal Large Language Models via Dynamic Visual-Token Exit and the Empirical Findings [66.04061083611863]
  Excessive use of visual tokens in existing Multimoal Large Language Models (MLLMs) often exhibits obvious redundancy and brings in prohibitively expensive computation.<n>We propose a simple yet effective method to improve the efficiency of MLLMs, termed dynamic visual-token exit (DyVTE)<n>DyVTE uses lightweight hyper-networks to perceive the text token status and decide the removal of all visual tokens after a certain layer.
  arXiv  Detail & Related papers  (2024-11-29T11:24:23Z)
• Get my drift? Catching LLM Task Drift with Activation Deltas [55.75645403965326]
  Task drift allows attackers to exfiltrate data or influence the LLM's output for other users.<n>We show that a simple linear classifier can detect drift with near-perfect ROC AUC on an out-of-distribution test set.<n>We observe that this approach generalizes surprisingly well to unseen task domains, such as prompt injections, jailbreaks, and malicious instructions.
  arXiv  Detail & Related papers  (2024-06-02T16:53:21Z)
• One Token Can Help! Learning Scalable and Pluggable Virtual Tokens for Retrieval-Augmented Large Language Models [67.49462724595445]
  Retrieval-augmented generation (RAG) is a promising way to improve large language models (LLMs)<n>We propose a novel method that involves learning scalable and pluggable virtual tokens for RAG.
  arXiv  Detail & Related papers  (2024-05-30T03:44:54Z)
• ST-LLM: Large Language Models Are Effective Temporal Learners [58.79456373423189]
  Large Language Models (LLMs) have showcased impressive capabilities in text comprehension and generation. How to effectively encode and understand videos in video-based dialogue systems remains to be solved. We propose ST-LLM, an effective video-LLM baseline with spatial-temporal sequence modeling inside LLM.
  arXiv  Detail & Related papers  (2024-03-30T10:11:26Z)
• u-LLaVA: Unifying Multi-Modal Tasks via Large Language Model [17.3535277338312]
  u-LLaVA is an innovative unifying multi-task framework that integrates pixel, regional, and global features to refine the perceptual faculties of MLLMs. This work contributes a novel mask-based multi-task dataset comprising 277K samples, crafted to challenge and assess the fine-grained perception capabilities of MLLMs.
  arXiv  Detail & Related papers  (2023-11-09T13:18:27Z)

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.