Alert button
Picture for Soumya Ghosh

Soumya Ghosh

Alert button

Multi-Level Explanations for Generative Language Models

Mar 21, 2024
Lucas Monteiro Paes, Dennis Wei, Hyo Jin Do, Hendrik Strobelt, Ronny Luss, Amit Dhurandhar, Manish Nagireddy, Karthikeyan Natesan Ramamurthy, Prasanna Sattigeri, Werner Geyer, Soumya Ghosh

Viaarxiv icon

Detectors for Safe and Reliable LLMs: Implementations, Uses, and Limitations

Mar 09, 2024
Swapnaja Achintalwar, Adriana Alvarado Garcia, Ateret Anaby-Tavor, Ioana Baldini, Sara E. Berger, Bishwaranjan Bhattacharjee, Djallel Bouneffouf, Subhajit Chaudhury, Pin-Yu Chen, Lamogha Chiazor, Elizabeth M. Daly, Rogério Abreu de Paula, Pierre Dognin, Eitan Farchi, Soumya Ghosh, Michael Hind, Raya Horesh, George Kour, Ja Young Lee, Erik Miehling, Keerthiram Murugesan, Manish Nagireddy, Inkit Padhi, David Piorkowski, Ambrish Rawat, Orna Raz, Prasanna Sattigeri, Hendrik Strobelt, Sarathkrishna Swaminathan, Christoph Tillmann, Aashka Trivedi, Kush R. Varshney, Dennis Wei, Shalisha Witherspooon, Marcel Zalmanovici

Viaarxiv icon

Improved Evidential Deep Learning via a Mixture of Dirichlet Distributions

Feb 09, 2024
J. Jon Ryu, Maohao Shen, Soumya Ghosh, Yuheng Bu, Prasanna Sattigeri, Subhro Das, Gregory W. Wornell

Viaarxiv icon

Assessment of Prediction Intervals Using Uncertainty Characteristics Curves

Oct 04, 2023
Jiri Navratil, Benjamin Elder, Matthew Arnold, Soumya Ghosh, Prasanna Sattigeri

Figure 1 for Assessment of Prediction Intervals Using Uncertainty Characteristics Curves
Figure 2 for Assessment of Prediction Intervals Using Uncertainty Characteristics Curves
Figure 3 for Assessment of Prediction Intervals Using Uncertainty Characteristics Curves
Figure 4 for Assessment of Prediction Intervals Using Uncertainty Characteristics Curves
Viaarxiv icon

Reliable Gradient-free and Likelihood-free Prompt Tuning

Apr 30, 2023
Maohao Shen, Soumya Ghosh, Prasanna Sattigeri, Subhro Das, Yuheng Bu, Gregory Wornell

Figure 1 for Reliable Gradient-free and Likelihood-free Prompt Tuning
Figure 2 for Reliable Gradient-free and Likelihood-free Prompt Tuning
Figure 3 for Reliable Gradient-free and Likelihood-free Prompt Tuning
Figure 4 for Reliable Gradient-free and Likelihood-free Prompt Tuning
Viaarxiv icon

Post-hoc Uncertainty Learning using a Dirichlet Meta-Model

Dec 14, 2022
Maohao Shen, Yuheng Bu, Prasanna Sattigeri, Soumya Ghosh, Subhro Das, Gregory Wornell

Figure 1 for Post-hoc Uncertainty Learning using a Dirichlet Meta-Model
Figure 2 for Post-hoc Uncertainty Learning using a Dirichlet Meta-Model
Figure 3 for Post-hoc Uncertainty Learning using a Dirichlet Meta-Model
Figure 4 for Post-hoc Uncertainty Learning using a Dirichlet Meta-Model
Viaarxiv icon

Fair Infinitesimal Jackknife: Mitigating the Influence of Biased Training Data Points Without Refitting

Dec 13, 2022
Prasanna Sattigeri, Soumya Ghosh, Inkit Padhi, Pierre Dognin, Kush R. Varshney

Figure 1 for Fair Infinitesimal Jackknife: Mitigating the Influence of Biased Training Data Points Without Refitting
Figure 2 for Fair Infinitesimal Jackknife: Mitigating the Influence of Biased Training Data Points Without Refitting
Figure 3 for Fair Infinitesimal Jackknife: Mitigating the Influence of Biased Training Data Points Without Refitting
Figure 4 for Fair Infinitesimal Jackknife: Mitigating the Influence of Biased Training Data Points Without Refitting
Viaarxiv icon

Are you using test log-likelihood correctly?

Dec 01, 2022
Sameer K. Deshpande, Soumya Ghosh, Tin D. Nguyen, Tamara Broderick

Figure 1 for Are you using test log-likelihood correctly?
Figure 2 for Are you using test log-likelihood correctly?
Figure 3 for Are you using test log-likelihood correctly?
Figure 4 for Are you using test log-likelihood correctly?
Viaarxiv icon

Post-hoc loss-calibration for Bayesian neural networks

Jun 13, 2021
Meet P. Vadera, Soumya Ghosh, Kenney Ng, Benjamin M. Marlin

Figure 1 for Post-hoc loss-calibration for Bayesian neural networks
Figure 2 for Post-hoc loss-calibration for Bayesian neural networks
Figure 3 for Post-hoc loss-calibration for Bayesian neural networks
Figure 4 for Post-hoc loss-calibration for Bayesian neural networks
Viaarxiv icon

Measuring the sensitivity of Gaussian processes to kernel choice

Jun 11, 2021
William T. Stephenson, Soumya Ghosh, Tin D. Nguyen, Mikhail Yurochkin, Sameer K. Deshpande, Tamara Broderick

Figure 1 for Measuring the sensitivity of Gaussian processes to kernel choice
Figure 2 for Measuring the sensitivity of Gaussian processes to kernel choice
Figure 3 for Measuring the sensitivity of Gaussian processes to kernel choice
Figure 4 for Measuring the sensitivity of Gaussian processes to kernel choice
Viaarxiv icon