Search Results - "Tamkin, Alex"

Codebook Features: Sparse and Discrete Interpretability for Neural Networks by Tamkin, Alex, Taufeeque, Mohammad, Goodman, Noah D

“…Understanding neural networks is challenging in part because of the dense, continuous nature of their hidden states. We explore whether we can train neural…”
Get full text

Operationalising the Definition of General Purpose AI Systems: Assessing Four Approaches by Uuk, Risto, Gutierrez, Carlos Ignacio, Tamkin, Alex

“…The European Union's Artificial Intelligence (AI) Act is set to be a landmark legal instrument for regulating AI technology. While stakeholders have primarily…”
Get full text

Multispectral Contrastive Learning with Viewmaker Networks by Bayrooti, Jasmine, Goodman, Noah, Tamkin, Alex

“…Contrastive learning methods have been applied to a range of domains and modalities by training models to identify similar "views" of data points. However,…”
Get full text

Multispectral Contrastive Learning with Viewmaker Networks by Bayrooti, Jasmine, Goodman, Noah, Tamkin, Alex

“…Contrastive learning methods have been applied to a range of domains and modalities by training models to identify similar "views" of data points. However,…”
Get full text

Oolong: Investigating What Makes Transfer Learning Hard with Controlled Studies by Wu, Zhengxuan, Tamkin, Alex, Papadimitriou, Isabel

“…When we transfer a pretrained language model to a new language, there are many axes of variation that change at once. To disentangle the impact of different…”
Get full text

Eliciting Human Preferences with Language Models by Li, Belinda Z, Tamkin, Alex, Goodman, Noah, Andreas, Jacob

“…Language models (LMs) can be directed to perform target tasks by using labeled examples or natural language prompts. But selecting examples or writing prompts…”
Get full text

Drone.io: A Gestural and Visual Interface for Human-Drone Interaction by Cauchard, Jessica R., Tamkin, Alex, Wang, Cheng Yao, Vink, Luke, Park, Michelle, Fang, Tommy, Landay, James A.

“…Drones are becoming ubiquitous and offer support to people in various tasks, such as photography, in increasingly interactive social contexts. We introduce…”
Get full text

Task Ambiguity in Humans and Language Models by Tamkin, Alex, Handa, Kunal, Shrestha, Avash, Goodman, Noah

“…Language models have recently achieved strong performance across a wide range of NLP benchmarks. However, unlike benchmarks, real world tasks are often poorly…”
Get full text

Feature Dropout: Revisiting the Role of Augmentations in Contrastive Learning by Tamkin, Alex, Glasgow, Margalit, He, Xiluo, Goodman, Noah

“…What role do augmentations play in contrastive learning? Recent work suggests that good augmentations are label-preserving with respect to a specific…”
Get full text

Collective Constitutional AI: Aligning a Language Model with Public Input by Huang, Saffron, Siddarth, Divya, Lovitt, Liane, Liao, Thomas I, Durmus, Esin, Tamkin, Alex, Ganguli, Deep

“…Proceedings of the 2024 ACM Conference on Fairness, Accountability, and Transparency. 1395-1417 There is growing consensus that language model (LM) developers…”
Get full text

Language Through a Prism: A Spectral Approach for Multiscale Language Representations by Tamkin, Alex, Jurafsky, Dan, Goodman, Noah

“…Language exhibits structure at different scales, ranging from subwords to words, sentences, paragraphs, and documents. To what extent do deep models capture…”
Get full text

Bayesian Preference Elicitation with Language Models by Handa, Kunal, Gal, Yarin, Pavlick, Ellie, Goodman, Noah, Andreas, Jacob, Tamkin, Alex, Li, Belinda Z

“…Aligning AI systems to users' interests requires understanding and incorporating humans' complex values and preferences. Recently, language models (LMs) have…”
Get full text

Viewmaker Networks: Learning Views for Unsupervised Representation Learning by Tamkin, Alex, Wu, Mike, Goodman, Noah

“…Many recent methods for unsupervised representation learning train models to be invariant to different "views," or distorted versions of an input. However,…”
Get full text

Active Learning Helps Pretrained Models Learn the Intended Task by Tamkin, Alex, Nguyen, Dat, Deshpande, Salil, Mu, Jesse, Goodman, Noah

“…Models can fail in unpredictable ways during deployment due to task ambiguity, when multiple behaviors are consistent with the provided training data. An…”
Get full text

Tradeoffs Between Contrastive and Supervised Learning: An Empirical Study by Karthik, Ananya, Wu, Mike, Goodman, Noah, Tamkin, Alex

“…Contrastive learning has made considerable progress in computer vision, outperforming supervised pretraining on a range of downstream datasets. However, is…”
Get full text

C5T5: Controllable Generation of Organic Molecules with Transformers by Rothchild, Daniel, Tamkin, Alex, Yu, Julie, Misra, Ujval, Gonzalez, Joseph

“…Methods for designing organic materials with desired properties have high potential impact across fields such as medicine, renewable energy, petrochemical…”
Get full text

Evaluating and Mitigating Discrimination in Language Model Decisions by Tamkin, Alex, Askell, Amanda, Lovitt, Liane, Durmus, Esin, Joseph, Nicholas, Kravec, Shauna, Nguyen, Karina, Kaplan, Jared, Ganguli, Deep

“…As language models (LMs) advance, interest is growing in applying them to high-stakes societal decisions, such as determining financing or housing eligibility…”
Get full text

Social Contract AI: Aligning AI Assistants with Implicit Group Norms by Fränken, Jan-Philipp, Kwok, Sam, Ye, Peixuan, Gandhi, Kanishk, Arumugam, Dilip, Moore, Jared, Tamkin, Alex, Gerstenberg, Tobias, Goodman, Noah D

“…We explore the idea of aligning an AI assistant by inverting a model of users' (unknown) preferences from observed interactions. To validate our proposal, we…”
Get full text

Understanding the Capabilities, Limitations, and Societal Impact of Large Language Models by Tamkin, Alex, Brundage, Miles, Clark, Jack, Ganguli, Deep

“…On October 14th, 2020, researchers from OpenAI, the Stanford Institute for Human-Centered Artificial Intelligence, and other universities convened to discuss…”
Get full text

Investigating Transferability in Pretrained Language Models by Tamkin, Alex, Singh, Trisha, Giovanardi, Davide, Goodman, Noah

“…How does language model pretraining help transfer learning? We consider a simple ablation technique for determining the impact of each pretrained layer on…”
Get full text