Token-Guard: Towards Token-Level Hallucination Control via Self-Checking Decoding

• URL: http://arxiv.org/abs/2601.21969v2
• Date: Fri, 30 Jan 2026 02:59:30 GMT
• Title: Token-Guard: Towards Token-Level Hallucination Control via Self-Checking Decoding
• Abstract summary: Token-Guard is a token-level hallucination control method based on self-checking decoding.<n> Experiments on HALU datasets show Token-Guard substantially reduces hallucinations and improves generation accuracy.
• Score: 8.629393350514972
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: Large Language Models (LLMs) often hallucinate, generating content inconsistent with the input. Retrieval-Augmented Generation (RAG) and Reinforcement Learning with Human Feedback (RLHF) can mitigate hallucinations but require resource-intensive retrieval or large-scale fine-tuning. Decoding-based methods are lighter yet lack explicit hallucination control. To address this, we present Token-Guard, a token-level hallucination control method based on self-checking decoding. Token-Guard performs internal verification at each reasoning step to detect hallucinated tokens before they propagate. Candidate fragments are further evaluated in a latent space with explicit hallucination risk scoring, while iterative pruning and regeneration dynamically correct detected errors. Experiments on HALU datasets show Token-Guard substantially reduces hallucinations and improves generation accuracy, offering a scalable, modular solution for reliable LLM outputs. Our code is publicly available.

Related papers

• Leveraging Low-Level Symbolic Competences for Unsupervised Grounding in Hallucination Detection [6.17293763300421]
  Hallucination detection is difficult due to opaque reasoning processes of language models.<n>In this work, we investigate whether an LLM can use an alternative, low level, symbolic competence such assql for unsupervised hallucination detection.<n>We find that our approach improves on direct prediction and competes with state-of-the-art hallucination detection methods.
  arXiv  Detail & Related papers  (2026-09-04T11:44:11Z)
• UniProbe: A Learnable Token-Level Hallucination Detector for Large VLMs using Multi-Structural Internal Representations [61.85916280854257]
  Large Vision-Language Models (LVLMs) achieve impressive visual reasoning and dialogue capabilities, yet frequently hallucinate content unsupported by the visual input.<n>We introduce textbfUniProbe, a lightweight, unified, learnable detector that models a frozen LVLM's heterogeneous computational trace from a single forward pass.
  arXiv  Detail & Related papers  (2026-08-11T12:01:59Z)
• VGS-Decoding: Visual Grounding Score Guided Decoding for Hallucination Mitigation in Medical VLMs [29.648561546849766]
  We propose Visual Grounding Score Guided Decoding (VGS-Decoding) to mitigate hallucinations during inference.<n>Our key insight is that hallucinated tokens maintain or increase their probability when visual information is degraded.<n>During decoding, we reweight probabilities by amplifying visually grounded tokens while suppressing hallucinations.
  arXiv  Detail & Related papers  (2026-03-19T18:26:23Z)
• Hallucination Begins Where Saliency Drops [18.189047289404325]
  hallucinations frequently arise when preceding output tokens exhibit low saliency toward the prediction of the next token.<n>We introduce LVLMs-Saliency, a gradient-aware diagnostic framework that quantifies the visual grounding strength of each output token.<n>Our method significantly reduces hallucination rates while preserving fluency and task performance, offering a robust and interpretable solution.
  arXiv  Detail & Related papers  (2026-01-28T05:50:52Z)
• DHI: Leveraging Diverse Hallucination Induction for Enhanced Contrastive Factuality Control in Large Language Models [33.2779808039684]
  Large language models (LLMs) frequently produce inaccurate or fabricated information, known as "hallucinations"<n>We propose DHI (Diverse Hallucination Induction), a novel training framework that generates a broader range of hallucinations without relying on pre-annotated data.<n> DHI achieves significant performance gains over other contrastive decoding-based approaches across multiple hallucination benchmarks.
  arXiv  Detail & Related papers  (2026-01-03T10:55:41Z)
• Steering LVLMs via Sparse Autoencoder for Hallucination Mitigation [38.43656456659151]
  Large vision-language models (LVLMs) have achieved remarkable performance on multimodal tasks.<n>They still suffer from hallucinations, generating text inconsistent with visual input, posing significant risks in real-world applications.<n>We propose Steering LVLMs via SAE Latent Directions (SSL), a plug-and-play method based on SAE-derived latent directions to mitigate hallucinations in LVLMs.
  arXiv  Detail & Related papers  (2025-05-22T02:45:45Z)
• HalluLens: LLM Hallucination Benchmark [49.170128733508335]
  Large language models (LLMs) often generate responses that deviate from user input or training data, a phenomenon known as "hallucination"<n>This paper introduces a comprehensive hallucination benchmark, incorporating both new extrinsic and existing intrinsic evaluation tasks.
  arXiv  Detail & Related papers  (2025-04-24T13:40:27Z)
• Generate, but Verify: Reducing Hallucination in Vision-Language Models with Retrospective Resampling [78.78822033285938]
  Vision-Language Models (VLMs) excel at visual understanding but often suffer from visual hallucinations.<n>In this work, we introduce REVERSE, a unified framework that integrates hallucination-aware training with on-the-fly self-verification.
  arXiv  Detail & Related papers  (2025-04-17T17:59:22Z)
• Attention Hijackers: Detect and Disentangle Attention Hijacking in LVLMs for Hallucination Mitigation [123.54980913741828]
  Large Vision-Language Models (LVLMs) remain vulnerable to hallucinations.<n>We propose a novel, training-free strategy namely Attention HIjackers Detection and Disentanglement (AID)<n>AID identifies Attention Hijackers by calculating instruction-driven visual salience.<n>Next, Attention Disentanglement mechanism is proposed to mask the visual attention of these identified Hijackers.<n>Re-Disentanglement recalculates the balance between instruction-driven and image-driven visual salience to avoid over-masking effects.
  arXiv  Detail & Related papers  (2025-03-11T09:35:55Z)
• Mitigating Entity-Level Hallucination in Large Language Models [11.872916697604278]
  This paper proposes Dynamic Retrieval Augmentation based on hallucination Detection (DRAD) as a novel method to detect and mitigate hallucinations in Large Language Models (LLMs) Experiment results show that DRAD demonstrates superior performance in both detecting and mitigating hallucinations in LLMs.
  arXiv  Detail & Related papers  (2024-07-12T16:47:34Z)
• Detecting and Mitigating Hallucination in Large Vision Language Models via Fine-Grained AI Feedback [40.930238150365795]
  We propose detecting and mitigating hallucinations in Large Vision Language Models (LVLMs) via fine-grained AI feedback.<n>We generate a small-size hallucination annotation dataset by proprietary models.<n>Then, we propose a detect-then-rewrite pipeline to automatically construct preference dataset for training hallucination mitigating model.
  arXiv  Detail & Related papers  (2024-04-22T14:46:10Z)
• OPERA: Alleviating Hallucination in Multi-Modal Large Language Models via Over-Trust Penalty and Retrospection-Allocation [124.9008419182485]
  We present OPERA, a novel MLLM decoding method grounded in an Over-trust Penalty and a Retrospection-Allocation strategy. Our approach begins with an interesting observation that, most hallucinations are closely tied to the knowledge aggregation patterns in the self-attention matrix. Based on the observation, OPERA introduces a penalty term on the model logits during the beam-search decoding to mitigate the over-trust issue.
  arXiv  Detail & Related papers  (2023-11-29T18:57:07Z)
• HalluciDoctor: Mitigating Hallucinatory Toxicity in Visual Instruction Data [102.56792377624927]
  hallucinations inherent in machine-generated data remain under-explored. We present a novel hallucination detection and elimination framework, HalluciDoctor, based on the cross-checking paradigm. Our method successfully mitigates 44.6% hallucinations relatively and maintains competitive performance compared to LLaVA.
  arXiv  Detail & Related papers  (2023-11-22T04:52:58Z)
• Detecting Hallucinated Content in Conditional Neural Sequence Generation [165.68948078624499]
  We propose a task to predict whether each token in the output sequence is hallucinated (not contained in the input) We also introduce a method for learning to detect hallucinations using pretrained language models fine tuned on synthetic data.
  arXiv  Detail & Related papers  (2020-11-05T00:18:53Z)

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.