DEEPO: Dual-Entropy Enhanced Policy Optimization for Hallucination in MLLMs
cs.AI, cs.LG
Submitted: 2026-09-23
Updated: 2026-09-23
Terminology
Sources
- SRFT: A Single-Stage Method with Supervised and Reinforcement Fine-Tuning for Reasoning
- Qwen2.5-VL Technical Report
- Hallucination of Multimodal Large Language Models: A Survey
- HalluLens: LLM Hallucination Benchmark
- Expanding Performance Boundaries of Open-Source Multimodal Models with Model, Data, and Test-Time Scaling
- Multi-Object Hallucination in Vision-Language Models
- Mitigating Hallucination in Visual Language Models with Visual Supervision
- A comprehensive taxonomy of hallucinations in Large Language Models
- Perception-Aware Policy Optimization for Multimodal Reasoning
- Seeing is Believing: Mitigating Hallucination in Large Vision-Language Models via CLIP-Guided Decoding
- Systematic Reward Gap Optimization for Mitigating VLM Hallucinations
- Spotlight on Token Perception for Multimodal Reinforcement Learning
- Writing-Zero: Bridge the Gap Between Non-verifiable Tasks and Verifiable Rewards
- Instanton Density Operator in Lattice QCD from Higher Category Theory
- Why Language Models Hallucinate
- Mitigating Object Hallucinations in Large Vision-Language Models through Visual Contrastive Decoding
- Evaluating Object Hallucination in Large Vision-Language Models
- Feasibility of Federated Learning from Client Databases with Different Brain Diseases and MRI Modalities
- Enhancing Advanced Visual Reasoning Ability of Large Language Models
- MMBench: Is Your Multi-modal Model an All-around Player?
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection