works on
evaluation protocol 1feed-forward network gating 1inference-time intervention 1structured output generation 1tool-using agents 1

From the 1 of 1 linked paper with an AI index.

collaborators
Showing cs.CLShow all

1 paper · 1 filter

cs.CL2026

Amplitude-Only FFN Intervention for Tool-Structured LLM Inference Method: Gated Evaluation Protocol, and Cross-Model Empirical Results

Sheng Xu, Junhua Wang, Boyuan Huang +4

The paper proposes Amplitude Gating, an inference‑time method that adjusts FFN activation magnitudes to improve tool‑structured outputs of large language models without changing mo…