Verifiable Rewards Beyond Math and Code: Lightweight Corpus-Grounded Process Supervision for Factual Question Answering
Paper • 2605.29648 • Published • 10
02. Prompt engineering
-
-
Why Larger Models Learn More: Effects of Capacity, Interference, and Rare-Task Retention
Paper • 2605.29548 • Published • 11 -
Towards Verifiable Multimodal Deep Research: A Multi-Agent Harness for Interleaved Report Generation
Paper • 2605.29861 • Published • 16 -
COLLEAGUE.SKILL: Automated AI Skill Generation via Expert Knowledge Distillation
Paper • 2605.31264 • Published • 112 -
dMoE: dLLMs with Learnable Block Experts
Paper • 2605.30876 • Published • 37 -
On the Geometry of On-Policy Distillation
Paper • 2606.07082 • Published • 71 -
SWE-Explore: Benchmarking How Coding Agents Explore Repositories
Paper • 2606.07297 • Published • 112 -
Whisper Hallucination Detection and Mitigation via Hidden Representation Steering and Sparse AutoEncoders
Paper • 2606.07473 • Published • 15 -
DuMate-DeepResearch: An Auditable Multi-Agent System with Recursive Search and Rubric-Grounded Reasoning
Paper • 2606.07299 • Published • 6 -
Why Muon Outperforms Adam: A Curvature Perspective
Paper • 2606.04662 • Published • 10 -
Experience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory
Paper • 2606.09365 • Published • 3 -
Kwai Keye-VL-2.0 Technical Report
Paper • 2606.10651 • Published • 184 -
Role-Agent: Bootstrapping LLM Agents via Dual-Role Evolution
Paper • 2606.10917 • Published • 76 -
Quickest Detection of Hallucination Onset: Delay Bounds and Learned CUSUM Statistics
Paper • 2606.12476 • Published -
Smaller Models are Natural Explorers for Policy-Level Diversity in GRPO
Paper • 2605.30789 • Published • 22