MA8a2.36

Personalized Federated Uncertainty-Aware LLM Fine-Tuning

Haotian Xiang, University of Georgia, United States; Ping Xu, Georgia State University, United States; Zhi Tian, George Mason University, United States; Qin Lu, University of Georgia, United States
Session:
MA8a2: Graph Learning, Signal Processing, and Networked Applications Poster
Track:
Networks and Graphs
Location:
Merrill Hall
Presentation Time:
Mon, 26 Oct, 08:15 - 09:55 PT (UTC -7)
Session Co-Chairs:
Caio Netto, Johns Hopkins University and Geert Leus, TU Delft, Netherlands
Presentation
Discussion
Resources
No resources available.
Session MA8a2
MA8a2.21: Improving Noise Robustness in Frame-Potential-Based Graph Signal Sampling
Jana M. Hauer, Vincent P. Grande, Michael T. Schaub, RWTH Aachen University, Germany
MA8a2.22: Hyperbolic Embeddings under Unknown Edges
Federico Larroca, Paola Bermolen, Marcelo Fiori, Sofía Pérez Casulo, Facultad de Ingeniería, Universidad de la República, Uruguay
MA8a2.23: Beyond Depth and Width: Dendritic Architectural Contributions to Efficient Image Classification
Zeinab Ahmadi, Spyridon Chavlis, Panayiota Poirazi, Foundation for Research and Technology Hellas (FORTH), Greece
MA8a2.24: Weighted Community Detection for Signed Multilayer Brain Networks
Sema Athamnah, Selin Aviyente, Michigan State University, United States
MA8a2.25: INFERRING THE GRAPH STRUCTURE OF IMAGES FOR GRAPH NEURAL NETWORKS
Mayur S. Gowda, John Shi, Carnegie Mellon University, United States; Augusto Santos, Instituto de Telecomunicações-IT, Portugal; José M. F. Moura, Carnegie Mellon University, United States
MA8a2.26: On the Robustness of Geometric Graph Neural Networks to Manifold Perturbations
Caio Deberaldini Netto, Joshua Brown, Luana Ruiz, Johns Hopkins University, United States
MA8a2.27: Adaptive Safety Policies for AMR Fleets: A Benchmark Study of the Safety-Delay Trade-off\\in Human-Shared Workspaces
Jiachen Li, Shihao Li, Jian Chu, Soovadeep Bakshi, Dongmei Chen, UT Austin, United States
MA8a2.28: Differentially Private Synthetic Voltage Phasor Release via Power-Flow Propagation of Loads
Andrew Campbell, Chenyue ZHANG, Anna Scaglione, Cornell University, United States; Sean Peisert, Lawrence Berkeley National Laboratory, United States
MA8a2.29: GeomFuse: Geometry-Driven Label Fusion using LLM Embeddings
Ecem Bozkurt, Antonio Ortega, University of Southern California, United States
MA8a2.30: Sparse Graph Learning from Sparse Data via Fiedler Number Maximization
Bahar OveisGharan, Gene Cheung, Andrew Eckford, York University, Canada
MA8a2.31: Iterative Message Passing Algorithms for Combined Team Formation and Assignment Problems
Alexios Rustom, University of Southern California, United States; Van Hovenga, University of California, Berkeley, United States; Thomas R. Halford, Caliola Engineering, United States; Keith M. Chugg, University of Southern California, United States
MA8a2.32: Low-Distortion Adaptive Spline Companding with Rational Peak Shaping for PAPR Reduction in OTFS Modulation
Svetislav Maric, UC San Diego, United States; Lazar Velimirovic, Serbian Academy of Sciences and Arts, Yugoslavia
MA8a2.33: Distributed Covariance-Agnostic Latent State Estimation with Consensus: A Model-Based Deep Learning Approach
George Stamatelis, Kyriakos Stylianopoulos, George Alexandropoulos, National and Kapodistrian University of Athens, Greece
MA8a2.34: SchedPriv: A Randomized Scheduling approach to Model Privacy in Federated Learning
Nomaan Kherani, Urbashi Mitra, University of Southern California, United States
MA8a2.35: Game-Theoretic Federated Aggregation for 6G Edge Intelligence Under Byzantine Attacks
Farhan Hamid, Nicolò Michelusi, Arizona State University, United States
MA8a2.36: Personalized Federated Uncertainty-Aware LLM Fine-Tuning
Haotian Xiang, University of Georgia, United States; Ping Xu, Georgia State University, United States; Zhi Tian, George Mason University, United States; Qin Lu, University of Georgia, United States
MA8a2.37: A Line-search-free Method for Adaptive Decentralized Optimization
Xiaokai Chen, Purdue University, United States; Ilya Kuruzov, Innopolis University, Russian Federation; Gesualdo Scutari, Purdue University, United States
MA8a2.38: Learning to Preserve Fisher Information for Estimation-Oriented Fixed Bit-Budget Compression
Amir Weiss, Bar-Ilan University, Israel
Contacts