Jose I. Mestre

Identifiers

• name variant Jose I. Mestre 0.60 · backfill

Papers (10)

1. Detecting Atypical Clients in Federated Learning via Representation-Level Divergence cs.LG · 2026 · author #2
2. StableGrad: Backward Scale Control without Batch Normalization cs.LG · 2026 · author #1
3. FedOUI: OUI-Guided Client Weighting for Federated Aggregation cs.LG · 2026 · author #2
4. OUI as a Structural Observable: Towards an Activation-Centric View of Neural Network Training cs.LG · 2026 · author #2
5. OUIDecay: Adaptive Layer-wise Weight Decay for CNNs Using Online Activation Patterns cs.LG · 2026 · author #2
6. Refresh-Scaling the Memory of Balanced Adam cs.LG · 2026 · author #3
7. FedSQ: Optimized Weight Averaging via Fixed Gating cs.LG · 2026 · author #2
8. $\lambda$-GELU: Learning Gating Hardness for Controlled ReLU-ization in Deep Networks cs.LG · 2026 · author #3
9. Why Adam Works Better with $\beta_1 = \beta_2$: The Missing Gradient Scale Invariance Principle cs.LG · 2026 · author #3
10. GLAI: GreenLightningAI for Accelerated Training through Knowledge Decoupling cs.LG · 2025 · author #1

Mentions

• 2605.22266 #2 · arxiv_oai · confidence 0.70 Jose I. Mestre
• 2605.19856 #1 · arxiv_oai · confidence 0.70 Jose I. Mestre

Frequent Coauthors

• Alberto Fern\'andez-Hern\'andez 10 shared papers
• Cristian P\'erez-Corral 10 shared papers
• Manuel F. Dolz 10 shared papers
• Enrique S. Quintana-Ort\'i 9 shared papers
• Jose Duato 4 shared papers
• Enrique S. Quitana-Ort\'i 1 shared papers
• Jos\'e Duato 1 shared papers