Metraipolis › archive › dc_72cb3049dce5
DSPO: Diversity-aware Subjective Policy Optimization for Robust Emotional Reasoning — arXiv:2609.36775
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.36775
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-07T18:38:25.421Z
- SHA-256 of the archived text
8cd19ad1be62d0e3d789d3fabf42b633b79a0cc9f6b025c1c9131d344dc6653d- Stable URL of this page
https://metraipolis.com/documento/dc_72cb3049dce5
How to cite
DSPO: Diversity-aware Subjective Policy Optimization for Robust Emotional Reasoning — arXiv:2609.36775 — arXiv (en), https://arxiv.org/abs/2609.36775 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_72cb3049dce5 (Metraipolis archive dc_72cb3049dce5, sha256 8cd19ad1be62d0e3d789d3fabf42b633b79a0cc9f6b025c1c9131d344dc6653d)
BibTeX
@misc{dc72cb3049dce5,
title = {DSPO: Diversity-aware Subjective Policy Optimization for Robust Emotional Reasoning — arXiv:2609.36775},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.36775},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_72cb3049dce5; sha256 8cd19ad1be62d0e3d789d3fabf42b633b79a0cc9f6b025c1c9131d344dc6653d},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.36775. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Reinforcement Learning has significantly advanced the complex reasoning capabilities of MLLMs. However, prevailing RL algorithms suffer a severe failure in emotion reasoning tasks. These methods heavily rely on deterministic hard-label supervision and point-wise isolated evaluation, creating a fundamental gap with the inherently subjective and continuously distributed nature of human emotions. Furthermore, unlike explicit physical objects, emotional states are deeply implicit within visual cues. This abstract nature exacerbates visual hallucinations in MLLMs, leading to plausible yet ungrounded emotional evidence. To address these limitations, we propose Diversity-Aware Subjective Policy Optimization (DSPO), a reinforcement learning framework that jointly promotes subjective affective coverage and visual grounding. First, we construct a context-grounded emotional distribution prior in the VAD space by combining the lexical prior of the annotated emotion with image-specific contextual information. Based on this prior, we introduce a Distribution-Aligned Emotional Diversity Reward (DEDR), which measures the leave-one-out marginal contribution of each candidate emotion within a rollout. DEDR rewards candidates whose inclusion brings the predicted affective set closer to the context-grounded prior, thereby preserving plausible subjective interpretations without encouraging unconstrained dispersion. We further develop Counterfactual Visual Intervention Gating (CVIG), which masks the visual region highlighted in the reasoning process and uses the resulting candidate-wise probability changes to reduce the weights of interpretations unsupported by visual evidence. Extensive experiments demonstrate that DSPO achieves state-of-the-art performance across multiple public benchmarks, especially on the cross-domain performance, i.e., improving +10.8\% on average cross-domain accuracy than EMO-R3. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-30 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.36775 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-07T18:04:24.740Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).