Predictive Credit: Measuring What Scientific Explanations Add to Experimental Forecasts
Research
Research News
New papers, breakthroughs and academic AI research
ContractRL: Shielded Group-Relative Policy Optimization for Auditable Tool-Call Repair
Research
Fault-Tolerant Budget Conservation in Distributed Multi-Agent Delegation
Research
Signed Lexical Confidence for Risk-Calibrated Intent Routing
Research
Mathematical Transfer in LLMs Follows Reasoning Approach More Than Topic
Research
What Should an Agent Remember? Disentangling Retention from Retrieval in Bounded-Memory Evaluation
Research
JevSpawn: Adaptive Agentic Inference through Compositional Action Spaces
Research
Emergent Unfaithfulness: How Alignment Training Causes Language Models to Silently Override Task Faithfulness
Research
How Divergence Becomes Decision Flips in Compressed Language Models
Research
Certainty Is Not Just Correctness: Rethinking Token-Level Certainty in LLM Reasoning
Research
DuplexSpeechBench-Document Grounding: Benchmarking Document Grounding and Hallucinations in Voice Agents
Research
CAST: Cost-Aware Speculative Trees from One-Pass Block Drafters
Research