AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

HERO: Enhancing Multimodal Faithfulness via Dynamic Entropy-Aware Reinforcement Learning

Xiongfeng Yang, Qingan Zhang, Yuheng Zhang, Kun-Yu Lin, Jian-Fang Hu, Dongmei Jiang, WEISHI ZHENG

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Reinforcement learning for visual reasoning · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add entropy-aware weighting to GRPO (ablation)vanilla GRPO · THRONEAcc: 80.6 -> 82.0Improved
add dynamic offensive/defensive balancing (ablation)GRPO + entropy weighting · THRONEAcc: 82.0 -> 82.8Improved
add variance-gated sampling to GRPO (ablation)vanilla GRPO · THRONEAcc: 80.6 -> 80.6No clear effect
replace random sampling with variance gating at same data budget (ablation)GRPO · THRONEAcc: 79.2 -> 80.6Improved
train on random 60% of data instead of full data (ablation)GRPO · THRONEAcc: 80.6 -> 79.2Hurt
add HERO versus vanilla GRPO (7B)Qwen2.5-VL-7B · POPEAcc: 89.2 -> 90.8Improved
add HERO versus vanilla GRPO (3B)Qwen2.5-VL-3B · THRONEAcc: 82.2 -> 82.8Improved
HERO versus offline HA-DPO (7B)Qwen2.5-VL-7B · POPEAcc: 87.2 -> 90.8Improved
HERO versus vanilla GRPO on AMBER hallucination rate (7B)Qwen2.5-VL-7B · AMBER-GCHAIR: 6.0 -> 6.3Hurt
HERO versus vanilla GRPO on AMBER coverage (7B)Qwen2.5-VL-7B · AMBER-GCover: 59.0 -> 61.2Improved
tune with CoT promptingvanilla baseline · POPEAcc: 89.6 -> 84.4Hurt
tune with CoT data (Tuned-CoT)vanilla baseline · POPEAcc: 89.6 -> 85.8Hurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy