# Adversarial Orthogonal Disentanglement for LVLM Hallucination Mitigation

arXiv: https://arxiv.org/abs/2605.25377 · HTML: https://mlsanity.com/paper/2605.25377/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Hallucination mitigation in LVLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 32 of its 48 references that are in the map.
- First posted on arXiv 25 May 2026.

## Placed in

- Multimodal LLMs / vision-language models / [Hallucination mitigation in LVLMs](https://mlsanity.com/mllm/hallucination-mitigation-in-lvlms/)

## Builds on (most cited first, 30 of 32)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [A Style-Based Generator Architecture for Generative Adversarial Networks](https://mlsanity.com/paper/1812.04948/) (2018)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Domain-Adversarial Training of Neural Networks](https://mlsanity.com/paper/1505.07818/) (2015)
- [MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI](https://mlsanity.com/paper/2311.16502/) (2023)
- [InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models](https://mlsanity.com/paper/2504.10479/) (2025)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [Representation Engineering: A Top-Down Approach to AI Transparency](https://mlsanity.com/paper/2310.01405/) (2023)
- [Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations](https://mlsanity.com/paper/1811.12359/) (2018)
- [Are We on the Right Way for Evaluating Large Vision-Language Models?](https://mlsanity.com/paper/2403.20330/) (2024)
- [Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding](https://mlsanity.com/paper/2311.16922/) (2023)
- [Object Hallucination in Image Captioning](https://mlsanity.com/paper/1809.02156/) (2018)
- [OCRBench: On the Hidden Mystery of OCR in Large Multimodal Models](https://mlsanity.com/paper/2305.07895/) (2023)
- [Extracting Latent Steering Vectors from Pretrained Language Models](https://mlsanity.com/paper/2205.05124/) (2022)
- [Vision-Language Models in Remote Sensing: Current Progress and Future Trends](https://mlsanity.com/paper/2305.05726/) (2023)
- [Reducing hallucination in structured outputs via Retrieval-Augmented Generation](https://mlsanity.com/paper/2404.08189/) (2024)
- [Inverse Reinforcement Learning with Dynamic Reward Scaling for LLM Alignment](https://mlsanity.com/paper/2503.18991/) (2025)
- [PBI-Attack: Prior-Guided Bimodal Interactive Black-Box Jailbreak Attack for Toxicity Maximization](https://mlsanity.com/paper/2412.05892/) (2024)
- [VASparse: Towards Efficient Visual Hallucination Mitigation via Visual-Aware Token Sparsification](https://mlsanity.com/paper/2501.06553/) (2025)
- [Object Detection with Multimodal Large Vision-Language Models: An In-depth Review](https://mlsanity.com/paper/2508.19294/) (2025)
- [TruthPrInt: Mitigating Large Vision-Language Models Object Hallucination Via Latent Truthful-Guided Pre-Intervention](https://mlsanity.com/paper/2503.10602/) (2025)
- [Attention Hijackers: Detect and Disentangle Attention Hijacking in LVLMs for Hallucination Mitigation](https://mlsanity.com/paper/2503.08216/) (2025)
- [Mitigating Hallucination in Large Vision-Language Models through Aligning Attention Distribution to Information Flow](https://mlsanity.com/paper/2505.14257/) (2025)
- [VIB-Probe: Detecting and Mitigating Hallucinations in Vision-Language Models via Variational Information Bottleneck](https://mlsanity.com/paper/2601.05547/) (2026)
- [Mitigating Dialogue Hallucination for Large Vision Language Models via Adversarial Instruction Tuning](https://mlsanity.com/paper/2403.10492/) (2024)
- [EcoAlign: An Economically Rational Framework for Efficient LVLM Alignment](https://mlsanity.com/paper/2511.11301/) (2025)
- [Causal-LLaVA: Causal Disentanglement for Mitigating Hallucination in Multimodal Large Language Models](https://mlsanity.com/paper/2505.19474/) (2025)
- [Steering the Verifiability of Multimodal AI Hallucinations](https://mlsanity.com/paper/2604.06714/) (2026)

## Built on by (most cited first, 1 of 1)

- [Search-G1: Grounded Search Agents via Representation-Based Intrinsic Rewards](https://mlsanity.com/paper/2608.07531/) (2026)

## Abstract (arXiv)

Large Vision-Language Models (LVLMs) have advanced multimodal understanding, yet their reliability is limited by hallucination, where generated content conflicts with visual facts. Existing mitigation methods either rely on costly external interventions, such as instruction tuning and retrieval, or use internal mechanisms that remain limited by flawed attention weights and entangled hidden representations. We propose Adversarial Orthogonal Disentanglement (AOD), a latent geometric framework for mitigating LVLM hallucinations. AOD learns a hallucination-related direction through a minimax objective: a classifier concentrates hallucination signals into the projected component, while an adversary removes them from the orthogonal residual space via a Gradient Reversal Layer. The learned direction enables a training-free dual-forward-pass contrastive decoding strategy that suppresses hallucinations while preserving general capabilities. Experiments on three LVLMs across four hallucination and four utility benchmarks show that AOD consistently outperforms strong baselines. It improves POPE accuracy by over 6% on average, boosts AMBER by 6%, and maintains strong performance on utility tasks such as MMMU. Further analysis shows robust transfer across datasets, suggesting that AOD captures general hallucination-related biases rather than dataset-specific artifacts. Our source code and datasets are available at https://github.com/Hunter-Wrynn/AOD.