3 papers
cs.AI2026
KnowRL: Boosting LLM Reasoning via Reinforcement Learning with Minimal-Sufficient Knowledge Guidance
Linhao Yu, Tianmeng Yang, Siyu Ding +8
RLVR improves reasoning in large language models, but its effectiveness is often limited by severe reward sparsity on hard problems. Recent hint-based RL methods mitigate sparsity…
astro-ph.HE2026
An extreme particle accelerator powered by pulsar PSR J1849-0001
The LHAASO Collaboration
Pulsar wind nebulae (PWNe) are bubbles of relativistic particles, powered by the rotational energy loss of the central pulsars. The Crab Nebula, powered by the Milky Way's most ene…
cs.CL2025
Towards Understanding Multi-Task Learning (Generalization) of LLMs via Detecting and Exploring Task-Specific Neurons
Yongqi Leng, Deyi Xiong
While large language models (LLMs) have demonstrated superior multi-task capabilities, understanding the learning mechanisms behind this is still a challenging problem. In this pap…