{"slug": "reframe-observation-aligned-multi-teacher-on-policy-distillation-for-agentic", "title": "Reframe: Observation-Aligned Multi-Teacher On-Policy Distillation for Agentic Visual Reasoning", "summary": "A paper submitted to arXiv on 9 November 2026 introduces Reframe, a post-training framework for visual agents that aligns the observation state rather than only the output distribution. Reframe combines Ground-Drop-GRPO (GD-GRPO), which trains spatial and temporal specialists with evidence-aware rewards and stochastic removal of the original visual input, with Observation-Gated-MOPD (OG-MOPD), which replaces deficient tool actions with canonical ones mid-rollout and resumes distillation from the repaired observation. Across 14 image and video benchmarks, Reframe outperforms both Mix-RL and vanilla multi-teacher on-policy distillation, and ablations show GD-GRPO increases observation dependence and tool gain compared with outcome-based GRPO, with neither stage requiring a task-specific SFT cold start.", "body_md": "# Computer Science > Computer Vision and Pattern Recognition\n\n  [Submitted on 9 Nov 2026]\n\n# Title:Reframe: Observation-Aligned Multi-Teacher On-Policy Distillation for Agentic Visual Reasoning\n\n[View PDF](http://arxiv.org/pdf/2611.12886v1)\n\n[HTML (experimental)](https://arxiv.org/html/2611.12886v1)\n\nAbstract:Agentic reinforcement learning (RL) and multi-teacher on-policy distillation (MOPD) offer complementary routes to post-training. For visual agents, however, tool use presents two related challenges. Models may overlook useful tool-returned observations, while poorly grounded actions can limit the evidence available for subsequent distillation. We introduce Reframe, a post-training framework that treats the observation state, rather than only the output distribution, as an object of alignment. First, Ground-Drop-GRPO (GD-GRPO) trains spatial and temporal specialists with evidence-aware rewards and stochastic removal of the original visual input, encouraging them to acquire grounded evidence and use the returned observations. Second, Observation-Gated-MOPD (OG-MOPD) replaces deficient tool actions with canonical ones mid-rollout and resumes distillation from the repaired observation. The two stages complement each other, as learning to rely on tool-returned observations can help the policy benefit from observation repair. Neither stage requires a task-specific SFT cold start. Extensive experiments over 14 image and video benchmarks demonstrate that Reframe outperforms both Mix-RL and vanilla MOPD. Ablation results show that GD-GRPO increases observation dependence and tool gain compared with outcome-based GRPO.\n    \n\n### Additional Features\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer \n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers \n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps \n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations \n\n*(*[What are Smart Citations?](https://www.scite.ai/))\n# Code, Data and Media Associated with this Article\n\nalphaXiv \n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers \n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub \n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub \n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face \n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast \n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))\n# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower \n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender \n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))\n# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [**Learn more about arXivLabs**](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/reframe-observation-aligned-multi-teacher-on-policy-distillation-for-agentic", "canonical_source": "http://arxiv.org/abs/2611.12886v1", "published_at": "2026-11-10 15:26:52+00:00", "updated_at": "2026-11-10 15:52:54.738387+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "computer-vision", "ai-research", "ai-agents"], "entities": ["Reframe", "Ground-Drop-GRPO", "Observation-Gated-MOPD", "GRPO", "arXiv"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/reframe-observation-aligned-multi-teacher-on-policy-distillation-for-agentic", "markdown": "https://wpnews.pro/news/reframe-observation-aligned-multi-teacher-on-policy-distillation-for-agentic.md", "text": "https://wpnews.pro/news/reframe-observation-aligned-multi-teacher-on-policy-distillation-for-agentic.txt", "jsonld": "https://wpnews.pro/news/reframe-observation-aligned-multi-teacher-on-policy-distillation-for-agentic.jsonld"}}