A hypernetwork generates meta-gating parameters for SwiGLU blocks to let LLMs adapt their nonlinearity to arbitrary textual conditions, outperforming finetuning and meta-learning baselines with reasonable generalization to unseen cases.
Metagater: Fast learning of conditional channel gated networks via federated meta-learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
verdicts
UNVERDICTED 2representative citing papers
A PPO deep RL agent learns to optimize cell offsets in a Python-simulated 5G environment, improving throughput, fairness, latency, jitter, packet loss, and handover counts over rule-based and other learning baselines under mobility and uncertainty.
citing papers explorer
-
Learn-to-learn on Arbitrary Textual Conditioning: A Hypernetwork-Driven Meta-Gated LLM
A hypernetwork generates meta-gating parameters for SwiGLU blocks to let LLMs adapt their nonlinearity to arbitrary textual conditions, outperforming finetuning and meta-learning baselines with reasonable generalization to unseen cases.
-
Deep Reinforcement Learning Approach to QoSAware Load Balancing in 5G Cellular Networks under User Mobility and Observation Uncertainty
A PPO deep RL agent learns to optimize cell offsets in a Python-simulated 5G environment, improving throughput, fairness, latency, jitter, packet loss, and handover counts over rule-based and other learning baselines under mobility and uncertainty.