…LLM-Designed Training Environment for RL with Multi-Agent Reasoning Published on Jun 16 Submitted by Chen chao on Jun 18 LARK Lab@HKUST (GZ) Authors: , , , , Abstract A framework automates environment redesign…
…https://researchpod.app/episode/33a09e3a-e6a1-46e4-9015-ffb8a1e4b07b Generated automatically by ResearchPod — happy to take feedback from the authors.
…visual plausibility and world reasoning: videos can look convincing while failing dynamics, causality, or information preservation. We will release our benchmarks and evaluation toolkit to support community research on genuinely world-aware…
Papers arxiv:2606.09380 Reasoning Arena: Trace Tournaments When Verifiable Rewards Fall Short Published on Jun 8 Submitted by Han Zhou on Jun 9 Mistral AI_ Authors: Han Zhou , , , , Abstract Reasoning Arena…
…Benchmarking Multi-hop Visual Reasoning for Visual DeepSearch (2026) Towards On-Policy Data Evolution for Visual-Native Multimodal Deep Search Agents (2026) InterSketch: An Interleaved Reasoning Model with Self-correcting Visual Sketch…
Papers arxiv:2606.16700 Multi-Turn Reflective Masking Elicits Reasoning in Mask Diffusion Models Published on Jun 15 Submitted by Tianyi Zhou on Jun 22 Tianyi Lab Authors: , Yihan Bian , , , , Tianyi Zhou…
…Deepening Reasoning in Multimodal Search Agents (2026) DR-Venus: Towards Frontier Edge-Scale Deep Research Agents with Only 10K Open Data (2026) Are Tools Always Beneficial? Learning to Invoke Tools Adaptively for…
…While prior research demonstrates that scaling environment quantity improves RL performance, existing manual or individual construction methods suffer from linear scaling limits, thereby hindering scalable reasoning generalization. This paper introduces RACES (Recursive…
…follow-up trace-diagnostic research. View arXiv page View PDF Add to collection Community TL;DR: Standard safety evaluations are missing a massive chunk of how reasoning models actually fail. In this…
…These results indicate that Chinese surface realization , translation artifacts, and model-specific behavior jointly affect multilingual logical reasoning. Overall, ChLogic provides a useful stress test for the robustness of multilingual reasoning . View…