Research
MagicBeanStalk: Escaping the RLVR Zero Advantage Trap with Recursive Data Synthesis (opens in a new tab)
arXiv cs.AI ·
Your next cue,
from NOCUE.
Research
Mitigating Capability Degradation in LLM Self-Improvement via Representation Geometry Regularization (opens in a new tab)
arXiv cs.AI ·
Research
Beyond Semantic IDs: Finite-Beam Reachability and Continuous-Query Generation in Generative Recommendation (opens in a new tab)
arXiv cs.AI ·
Research
Know What You Don't Know: Agentic Data Synthesis by Evolving Generation Space for Closed-Loop Learning (opens in a new tab)
arXiv cs.AI ·
Research
Teaching neural decoders with language (opens in a new tab)
arXiv cs.AI ·
Research
EliteTable: Self-Improving Program Search for Executable Control Policies (opens in a new tab)
arXiv cs.AI ·
Research
Mind the Spatial Gap: A Benchmark for Geospatial Reasoning and Constrained Spatial Editing (opens in a new tab)
arXiv cs.AI ·
Research
CLEAT: Learning Value-Guided Clarification for Interactive Agents (opens in a new tab)
arXiv cs.AI ·
Research
Work2Skill: Capturing Human GUI Experience for Reusable Agent Skill Induction (opens in a new tab)
arXiv cs.AI ·