1 code implementation • 11 Jan 2024 • Asma Ghandeharioun, Avi Caciularu, Adam Pearce, Lucas Dixon, Mor Geva
We introduce a framework called Patchscopes and show how it can be used to answer a wide range of questions about an LLM's computation.
no code implementations • 6 Dec 2023 • Dan Friedman, Andrew Lampinen, Lucas Dixon, Danqi Chen, Asma Ghandeharioun
A common method to study deep learning systems is to use simplified model representations -- for example, using singular value decomposition to visualize the model's hidden states in a lower dimensional space.
no code implementations • NeurIPS 2023 • Satyapriya Krishna, Jiaqi Ma, Dylan Slack, Asma Ghandeharioun, Sameer Singh, Himabindu Lakkaraju
Large Language Models (LLMs) have demonstrated remarkable capabilities in performing complex tasks.
no code implementations • 24 Jan 2023 • Robert A. Lewis, Asma Ghandeharioun, Szymon Fedor, Paola Pedrelli, Rosalind Picard, David Mischoulon
We suggest that this improved performance results from the ability of the mixed effects random forest to personalise model parameters to individuals in the dataset.
1 code implementation • NeurIPS 2023 • Peter Hase, Mohit Bansal, Been Kim, Asma Ghandeharioun
This finding raises questions about how past work relies on Causal Tracing to select which model layers to edit.
1 code implementation • ICLR 2022 • Asma Ghandeharioun, Been Kim, Chun-Liang Li, Brendan Jou, Brian Eoff, Rosalind W. Picard
Explaining deep learning model inferences is a promising venue for scientific understanding, improving safety, uncovering hidden biases, evaluating fairness, and beyond, as argued by many scholars.
1 code implementation • EMNLP 2020 • Natasha Jaques, Judy Hanwen Shen, Asma Ghandeharioun, Craig Ferguson, Agata Lapedriza, Noah Jones, Shixiang Shane Gu, Rosalind Picard
We start by hosting models online, and gather human feedback from real-time, open-ended conversations, which we then use to train and improve the models using offline reinforcement learning (RL).
no code implementations • ICLR 2020 • Natasha Jaques, Asma Ghandeharioun, Judy Hanwen Shen, Craig Ferguson, Agata Lapedriza, Noah Jones, Shixiang Gu, Rosalind Picard
This is a critical shortcoming for applying RL to real-world problems where collecting data is expensive, and models must be tested offline before being deployed to interact with the environment -- e. g. systems that learn from human interaction.
1 code implementation • 20 Sep 2019 • Asma Ghandeharioun, Brian Eoff, Brendan Jou, Rosalind W. Picard
Supporting model interpretability for complex phenomena where annotators can legitimately disagree, such as emotion recognition, is a challenging machine learning task.
1 code implementation • 17 Sep 2019 • Abdelrhman Saleh, Natasha Jaques, Asma Ghandeharioun, Judy Hanwen Shen, Rosalind Picard
Open-domain dialog generation is a challenging problem; maximum likelihood training can lead to repetitive outputs, models have difficulty tracking long-term conversational goals, and training on standard movie or online datasets may lead to the generation of inappropriate, biased, or offensive text.
1 code implementation • 30 Jun 2019 • Natasha Jaques, Asma Ghandeharioun, Judy Hanwen Shen, Craig Ferguson, Agata Lapedriza, Noah Jones, Shixiang Gu, Rosalind Picard
Most deep reinforcement learning (RL) systems are not able to learn effectively from off-policy data, especially if they cannot explore online in the environment.
2 code implementations • NeurIPS 2019 • Asma Ghandeharioun, Judy Hanwen Shen, Natasha Jaques, Craig Ferguson, Noah Jones, Agata Lapedriza, Rosalind Picard
To investigate the strengths of this novel metric and interactive evaluation in comparison to state-of-the-art metrics and human evaluation of static conversations, we perform extended experiments with a set of models, including several that make novel improvements to recent hierarchical dialog generation architectures through sentiment and semantic knowledge distillation on the utterance level.
no code implementations • 17 Apr 2017 • Ardavan Saeedi, Matthew D. Hoffman, Stephen J. DiVerdi, Asma Ghandeharioun, Matthew J. Johnson, Ryan P. Adams
Professional-grade software applications are powerful but complicated$-$expert users can achieve impressive results, but novices often struggle to complete even basic tasks.