Amazon · 16 records

dateidtitleauthorsvenue↑
2026-10-192610.22448Lean4AgenticRL: Advancing Agentic RL with Verifiable Multi-level Process RewardsRuida Wang et al.arXiv24
2026-10-082609.30541AutoResearch at Production Scale: Failure Modes and a Multi-Agent FrameworkAparajith Chandran et al.arXiv3
2026-10-062610.04198ALoDLM: Adaptively Looped Diffusion Language ModelsLiancheng Fang et al.arXiv74
2026-10-012609.34563Rethinking Latent Visual Reasoning: Grounding Latent Reasoning in Visual EvidenceXi Xiao et al.arXiv28
2026-10-012609.38660Breaking Babel: A Self-Evolving Multi-Agent System for Long-Form Subtitle TranslationHaibo Jin et al.arXiv22
2026-10-012609.38349MILO: Automated Harness Discovery via Orchestrated Multi-Agent EvolutionPrithwish Jana et al.arXiv18
2026-10-012609.38792Training LLM Judges from Language Feedback via Position-Selective Self-DistillationIlgee Hong et al.arXiv5
2026-09-302609.38360On the Off-Policy Teacher in On-Policy DistillationLanglin Huang et al.arXiv23
2026-09-302609.36352StructRL: Online Structured Reinforcement Learning for Long-Horizon Vision-Language-Action TasksZiyi Yin et al.arXiv5
2026-09-252609.29421Rufus-Air: An Open LLM Post-Training RecipeChia-Yuan Chang et al.arXiv39
2026-09-182609.20715Don't Mask the Environment: Observation Supervision Changes How Agents Explore Under RLJuzheng Zhang et al.arXiv19
2026-09-152609.12419MInTRL: Off-policy Intervention can boost On-policy RLMingyu Chen et al.arXiv12
2026-09-072609.00444Group Adaptive Clipping Policy OptimizationSheng Jia et al.arXiv9
2026-08-202608.14929Training Leaves Traces: Centered Residual Signatures for Language Model Lineage VerificationAman Singh Thakur et al.arXiv7
2026-06-152606.05308Statistically Reliable LLM-Based Ranking Evaluation via Prediction-Powered InferenceAbhishek DivekararXiv2
2026-05-142605.11378An Empirical Study of Automating Agent EvaluationKang Zhou et al.arXiv2