JudgeMyAIGraphDancer: How Curriculum Reinforcement Learning Unlocks Cross-Domain Graph Reasoning in LLMs
GraphDancer leverages curriculum reinforcement learning to train LLMs for cross-domain graph reasoning.
Compaction Strategies in Long Horizon Agentic Tasks: A Comprehensive Analysis
Explore the impact of compaction strategies on long horizon agentic tasks with frontier models.
A-MapReduce: How Agentic MapReduce Achieves State-of-the-Art Wide Search Performance
A-MapReduce revolutionizes wide search with agentic MapReduce, delivering 5.11% to 17.50% F1 improvements and 45.8% faster execution.
SOPO: Socratic Guided Policy Optimization for Span-Level Hallucination Detection
SOPO: Socratic Guided Policy Optimization for Span-Level Hallucination Detection
Quant-Arena: Benchmarking LLM Agents for Quantitative Trading Research
Quant-Arena benchmarks LLM agents for quantitative trading research, evaluating end-to-end strategy development and execution.
CHIMERA Dataset Delivers 4B Model Reasoning Performance on Par with 235B Models
CHIMERA synthetic dataset enables 4B models to achieve reasoning performance comparable to 235B models.
Gnani-Evon-v3.3-30B-A3B: A 5.2-Point Indic Reasoning Gain at 1.7-Point English Cost
Analysis of Gnani-Evon-v3.3-30B-A3B’s 5.2-point Indic reasoning gain and 1.7-point English cost.
Global Cross-Region Inference: Unlocking OpenAI GPT-6 Models in Asia Pacific
Explore global cross-region inference for OpenAI GPT-6 models in Asia Pacific with Amazon Bedrock.
Mine Odyssey Benchmark Reveals 61.7% Spatial Agentic Intelligence Gap Between GPT-6 Astra and DeepSeek-V4.1-Flash
Mine Odyssey reveals 61.7% spatial agentic intelligence gap between GPT-6 Astra and DeepSeek-V4.1-Flash.
BFCL Benchmark Flaw: Why Asking Earns Nothing in Multi-Turn Dialogues
BFCL benchmark flaw: Why asking earns nothing in multi-turn dialogues.