…Minimal-Interface Zero-Shot Agents Rival Industrial-Scale Policies in Vision-and-Language Navigation (2026) Nanbeige4.2-3B: Unlocking Agentic Capabilities in a Compact Model (2026) RoboBRIDGE: A Modular Framework for Bridging…
…Efficient and Instant Agentic Intelligence at Trillion-Parameter Scale Published on Jun 13 Submitted by Adina Yakefu on Jun 16 inclusionAI Authors: , , , , , , , , , Changxin Tian , , , , , , , , , , , , Abstract Ling-2.6 and Ring-2.6…
…We study a harder and more policy-relevant failure mode: no hidden text, no prompt injection, and no changes to methods, experiments, figures, equations, proofs, or numerical results. The attacker modifies only…
…On-Policy Self-Distillation for Long-Context Language Models (2026) A Survey of On-Policy Distillation for Large Language Models (2026) TRIMS: Trajectory-Ranked Instruction Masked Supervision for Diffusion Language Models (2026…
…Generated by Qwen/Qwen2.5-Coder-32B-Instruct World models that capture how actions induce physical change enable scalable robot learning without reliance on embodiment-specific action labels. Pixel-space video models…
…We first model localized support interactions as Intervention Units (IUs), which capture state--action--outcome dynamics between seeker states, support interventions, and post-response emotional changes. Based on IUs extracted from both…
…Generated by Qwen/Qwen2.5-Coder-32B-Instruct Understanding and predicting how social beliefs evolve in response to events -- from policy changes to scientific breakthroughs -- remains a fundamental challenge in social science…
Papers arxiv:2606.12203 Adaptive Multi-Resolution Procedural Knowledge Compression for Large Language Models Published on Jun 10 Submitted by Changyue Wang on Jun 11 Authors: , , , , , , , Abstract SKIM is an adaptive multi…
…A coding assistant that remembers "last time I said 'fix the error handling' I meant add retries, not change the exception type" would save me more time than any model upgrade. The…
…each action changes what must be perceived, reasoned about, grounded, and verified next. This paper presents Capek 0.5, an embodied VLM that organizes post-training around four execution-facing capabilities and…