Amortized Off-Policy Evaluation for LLMs
Research
Research News
New papers, breakthroughs and academic AI research
ReCast: Attribution-Oriented Step Representation Learning for LLM-Based Agent Systems
Research
UniData: Universal Multimodal Instruction Generation Pipeline
Research
Rendering-Free Lookahead for Question-Guided Active Vision
Research
MedBenchAgent: Towards Systematic Automation of Medical VLM Benchmark Construction
Research
EvoSim: Learning to Model, Modeling to Learn
Research
SynCo: Data Synthesis Co-Training for Self-Evolving LLMs via Multi-Agent Reinforcement Learning
Research
RL-ARC: Calibrating Large Reasoning Models via Reasoning-guided Uncertainty
Research
AffordDrive3D: Affordance-Aware World-Action Modeling with Spatial Understanding
Research
Can a System-One LLM Perform Knowledge Tracing When Few or No Learners Are Logged?
Research
RAG-Stress: Probing the Limits of Evidence Reliance in Retrieval-Augmented Generation
Research
MotherTree: Meta-learning on synthetic data improves decision tree training
Research