{
  "title": "Daily Research Brief 2026-04-14",
  "url": "/en/posts/research-brief-2026-04-14/",
  "permalink": "https://hackcv.com/en/posts/research-brief-2026-04-14/",
  "date": "2026-04-14",
  "lastmod": "2026-04-14",
  "author": "",
  "description": "Daily research brief — AI / LLM / Agent / Computer Vision / Audio-Video / Engineering",
  "categories": ["Research Brief"],
  "tags": ["AI","LLM","Agent","Computer Vision","Audio-Video","Engineering","Daily Brief"],
  "cover": "https://picsum.photos/seed/daily-research-brief-2026-04-14/1200/675",
  "readingTime": 1,
  "wordCount": 112,
  "content": "\u003ch1 id=\"daily-research-brief-2026-04-14\"\u003eDaily Research Brief 2026-04-14\u003c/h1\u003e\n\u003cp\u003e📊 Token usage: estimated from retrieval and writing scale.\u003c/p\u003e\n\u003cp\u003eCovers the latest AI research, open source and industry moves, updated daily.\u003c/p\u003e\n\u003chr\u003e\n\u003ch2 id=\"editors-note\"\u003eEditor\u0026rsquo;s Note\u003c/h2\u003e\n\u003cp\u003eThe paper thread today: invariance in vision models, pair-based 3D scene generation, physics-informed state-space models, safety-violation detection across agent behaviors, and RL for physics olympiads.\u003c/p\u003e\n\u003ch2 id=\"1-latest-arxiv-papers\"\u003e1. Latest arXiv Papers\u003c/h2\u003e\n\u003col\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eWho Handles Orientation? Investigating Invariance in Vision Models\u003c/strong\u003e\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003ePair2Scene: Learning Local Object Relations for 3D Scene Generation\u003c/strong\u003e\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003ePhysics-Informed State Space Models for Reliable Dynamics\u003c/strong\u003e\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eDetecting Safety Violations Across Many Agent Behaviors\u003c/strong\u003e\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eSolving Physics Olympiad via Reinforcement Learning\u003c/strong\u003e\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eOmniShow: Unifying Multimodal Conditions for Human Animation\u003c/strong\u003e\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eBudget-Aware Uncertainty for Radiotherapy Segmentation\u003c/strong\u003e\u003c/p\u003e\n\u003c/li\u003e\n\u003cli\u003e\n\u003cp\u003e\u003cstrong\u003eSyncFix: Fixing 3D Reconstructions via Multi-View Consistency\u003c/strong\u003e\u003c/p\u003e\n\u003c/li\u003e\n\u003c/ol\u003e\n",
  "summary": "Daily Research Brief 2026-04-14 📊 Token usage: estimated from retrieval and writing scale.\nCovers the latest AI research, open source and industry moves, updated daily.\nEditor\u0026rsquo;s Note The paper thread today: invariance in vision models, pair-based 3D scene generation, physics-informed state-space models, safety-violation detection across agent behaviors, and RL for physics olympiads.\n1. Latest arXiv Papers Who Handles Orientation? Investigating Invariance in Vision Models\nPair2Scene: Learning Local Object Relations for 3D Scene Generation\n"
}