Overview

Contrastive decoding studies how comparing a model’s predictions under different or counterfactual conditions can strengthen useful evidence and suppress unwanted priors. Common approaches contrast positive and negative prompts, visual or audio perturbations, model layers, or internal decoding paths, often with adaptive gating that intervenes only on uncertain tokens. The methods are applied to hallucination reduction, grounded vision and speech understanding, factual question answering, reasoning, and behavior control, usually without retraining. Gains depend on the model, task, and kind of error, so a central challenge is improving grounding without overcorrecting or mistaking benchmark improvements for genuine reliability.

Latest papers

CardsList

56 papers6 papers in the last four weeks, up 20% on the four weeks before. 0.0% of all new papers.

  1. Do We Really Need External Tools to Mitigate Hallucinations? SIRA: Shared-Prefix Internal Reconstruction of Attribution

    May 14, 2026Tian Qin, Junzhe Chen, Yuqing Shi +3Efficient VLM InferenceHallucination in Language Models

  2. When Looking Is Not Enough: Visual Attention Structure Reveals Hallucination in MLLMs

    May 12, 2026Fanpu Cao, Xin Zou, Xuming Hu +1Hallucination Detection in VLMsHallucination in Language Models

  3. APCD: Adaptive Path-Contrastive Decoding for Reliable Large Language Model Generation

    May 10, 2026Tianyu Zheng, Hong Wu, Jiaji ZhongLanguage Model DecodingLLM Reliability

  4. Hallucination as an Anomaly: Dynamic Intervention via Probabilistic Circuits

    May 7, 2026Erik Nielsen, Elia Cunegatti, Marcus Vukojevic +1Probabilistic CircuitsLLM Hallucination Mitigation

  5. Focus on the Core: Empowering Diffusion Large Language Models by Self-Contrast

    May 2, 2026Jinyuan Feng, Xin Yu, Yiqun Chen +5Diffusion Language Model InferenceDiffusion Language Model Decoding

  6. Instruction-Evidence Contrastive Dual-Stream Decoding for Grounded Vision-Language Reasoning

    Apr 28, 2026Yashwant Pravinrao Bangde, Debaditya RoyVLM InferenceObject Hallucination in VLMs

  7. Global Context or Local Detail? Adaptive Visual Grounding for Hallucination Mitigation

    Apr 27, 2026Yubo Jiang, Xin Yang, Abudukelimu Wuerkaixi +7Hallucination in Language ModelsObject Hallucination in VLMs

  8. Breaking the Illusion: When Positive Meets Negative in Multimodal Decoding

    Apr 22, 2026Yubo Jiang, Yitong An, Xin Yang +7Vision-Language ModelsHallucination in Language Models

  9. CCTVBench: Contrastive Consistency Traffic VideoQA Benchmark for Multimodal LLMs

    Apr 22, 2026Xingcheng Zhou, Hao Guo, Rui Song +5Visual Question AnsweringVideo Question Answering

  10. Please refuse to answer me! Mitigating Over-Refusal in Large Language Models via Adaptive Contrastive Decoding

    Apr 18, 2026Yupeng Qi, Ziyu Lyu, Lixin Cui +2Language Model Safety EvaluationLLM Refusal Behavior

  11. Temporal Contrastive Decoding: A Training-Free Method for Large Audio-Language Models

    Apr 16, 2026Yanda Li, Yuhan Liu, Zirui Song +3Audio UnderstandingAudio-Language Models

  12. The Cost of Language: Centroid Erasure Exposes and Exploits Modal Competition in Multimodal Language Models

    Apr 15, 2026Akshay Paruchuri, Ishan Chatterjee, Henry Fuchs +2Modality Reliance in Multimodal ModelsMultimodal Large Language Models

  13. Synthesizing Instruction-Tuning Datasets with Contrastive Decoding

    Apr 15, 2026Tatsuya Ichinose, Youmi Ma, Masanari Oi +2Instruction TuningSynthetic Data Generation for Language Models

  14. CWCD: Category-Wise Contrastive Decoding for Structured Medical Report Generation

    Apr 12, 2026Shantam Srivastava, Mahesh Bhosale, David Doermann +1Radiology Report GenerationRadiology VLMs

  15. Spotlight and Shadow: Attention-Guided Dual-Anchor Introspective Decoding for MLLM Hallucination Mitigation

    Apr 11, 2026Yebo Wu, Han Jin, Zhijiang Guo +1VLM Hallucination MitigationLanguage Model Decoding

  16. 3D-VCD: Hallucination Mitigation in 3D-LLM Embodied Agents through Visual Contrastive Decoding

    Apr 9, 2026Makanjuola Ogunleye, Eman Abdelrahman, Ismini LourentzouVLM Hallucination MitigationObject Hallucination in VLMs

  17. BUZZY: Contrastive Scoring to Mitigate Text-Induced Bias in Multimodal Multiple-Choice QA

    Mar 30, 2026Taeyun Roh, Suhyeong Park, Dongyoung Lee +4Vision-Language ModelsMultimodal QA

  18. Revealing Multi-View Hallucination in Large Vision-Language Models

    Mar 25, 2026Wooje Park, Insu Lee, Soohyun Kim +4Visual Question AnsweringHallucination in Language Models

  19. How Contrastive Decoding Enhances Large Audio Language Models

    Mar 10, 2026Tzu-Quan Lin, Wei-Ping Huang, Yi-Cheng Lin +1Contrastive DecodingAudio-Language Models

  20. Scaling Test-Time Robustness of Vision-Language Models via Self-Critical Inference Framework

    Mar 8, 2026Kaihua Tang, Jiaxin Qi, Jinli Ou +2VLM RobustnessTest-Time Scaling for VLMs

  21. Whisper-CD: Accurate Long-Form Speech Recognition using Multi-Negative Contrastive Decoding

    Mar 6, 2026Hoseong Ahn, Jeongyun Chae, Yoonji Park +1Language Model DecodingAutomatic Speech Recognition

  22. ThinkOmni: Lifting Textual Reasoning to Omni-modal Scenarios via Guidance Decoding

    Feb 26, 2026Yiran Guan, Sifan Tu, Dingkang Liang +6Multimodal ReasoningMLLM Reasoning

  23. PromptCD: Test-Time Behavior Enhancement via Polarity-Prompt Contrastive Decoding

    Feb 24, 2026Baolong Bi, Yuyao Ge, Shenghua Liu +9Inference-Time Language Model AlignmentLLM Alignment

  24. Thinking by Subtraction: Confidence-Driven Contrastive Decoding for LLM Reasoning

    Feb 20, 2026Lexiang Tang, Weihao Gao, Bingchen Zhao +4Efficient Language Model ReasoningLLM Reliability

  25. MAD: Modality-Adaptive Decoding for Mitigating Cross-Modal Hallucinations in Multimodal Large Language Models

    Jan 29, 2026Sangyun Chung, Se Yeon Kim, Youngchae Chee +1Multimodal RobustnessHallucination in Language Models