Zobrazeno 1 - 10
of 1 435
pro vyhledávání: '"Hsu, Hsiang"'
LLM hallucination, where LLMs occasionally generate unfaithful text, poses significant challenges for their practical applications. Most existing detection methods rely on external knowledge, LLM fine-tuning, or hallucination-labeled datasets, and th
Externí odkaz:
http://arxiv.org/abs/2411.09689
The growing richness of large-scale datasets has been crucial in driving the rapid advancement and wide adoption of machine learning technologies. The massive collection and usage of data, however, pose an increasing risk for people's private and sen
Externí odkaz:
http://arxiv.org/abs/2405.14981
Recent works have shown that by using large pre-trained models along with learnable prompts, rehearsal-free methods for class-incremental learning (CIL) settings can achieve superior performance to prominent rehearsal-based ones. Rehearsal-free CIL m
Externí odkaz:
http://arxiv.org/abs/2402.04129
Machine unlearning has emerged as a new paradigm to deliberately forget data samples from a given model in order to adhere to stringent regulations. However, existing machine unlearning methods have been primarily focused on classification models, le
Externí odkaz:
http://arxiv.org/abs/2402.00351
Predictive multiplicity refers to the phenomenon in which classification tasks may admit multiple competing models that achieve almost-equally-optimal performance, yet generate conflicting outputs for individual samples. This presents significant con
Externí odkaz:
http://arxiv.org/abs/2402.00728
The rapid growth of machine learning has spurred legislative initiatives such as ``the Right to be Forgotten,'' allowing users to request data removal. In response, ``machine unlearning'' proposes the selective removal of unwanted data without the ne
Externí odkaz:
http://arxiv.org/abs/2312.14923
Autor:
Sterken, Veerle J., Hunziker, Silvan, Dialynas, Kostas, Leitner, Jan, Sommer, Maximilian, Srama, Ralf, Baalmann, Lennart R., Li, Aigen, Herbst, Konstantin, Galli, André, Brandt, Pontus, Riebe, My, Baggaley, Jack, Blanc, Michel, Czechowski, Andrej, Effenberger, Frederic, Fields, Brian, Frisch, Priscilla, Horanyi, Mihaly, Hsu, Hsiang-Wen, Khawaja, Nozair, Krüger, Harald, Kurth, Bill S., Ligterink, Niels F. W., Linsky, Jeffrey L., Lisse, Casey, Malaspina, David, Miller, Jesse A., Opher, Merav, Poppe, Andrew R., Postberg, Frank, Provornikova, Elena, Redfield, Seth, Richardson, John, Rowan-Robinson, Michael, Scherer, Klaus, Shen, Mitchell M., Slavin, Jon D., Sternovsky, Zoltan, Stober, Gunter, Strub, Peter, Szalay, Jamey, Trieloff, Mario
Publikováno v:
RAS Techniques and Instruments, rzad034 (2023)
We discuss the synergies between heliospheric and dust science, the open science questions, the technological endeavors and programmatic aspects that are important to maintain or develop in the decade to come. In particular, we illustrate how we can
Externí odkaz:
http://arxiv.org/abs/2308.10728
Machine learning tasks may admit multiple competing models that achieve similar performance yet produce conflicting outputs for individual samples -- a phenomenon known as predictive multiplicity. We demonstrate that fairness interventions in machine
Externí odkaz:
http://arxiv.org/abs/2306.09425
Space missions often carry antenna instruments that are sensitive to dust impacts, however, the understanding of signal generation mechanisms remained incomplete. A signal generation model in an analytical form is presented that provides a good agree
Externí odkaz:
http://arxiv.org/abs/2304.00453
Mechanisms used in privacy-preserving machine learning often aim to guarantee differential privacy (DP) during model training. Practical DP-ensuring training methods use randomization when fitting model parameters to privacy-sensitive data (e.g., add
Externí odkaz:
http://arxiv.org/abs/2302.14517