Dylan Sam

23 papers A* 5A 3Journal 15
YearRankTypeTitle / Venue / Authors
2026 J jnl
When Should We Introduce Safety Interventions During Pretraining?
CoRR
Dylan Sam, Sachin Goyal, Pratyush Maini, Alexander Robey, J. Zico Kolter
2025 J jnl
Analyzing Similarity Metrics for Data Selection for Language Model Pretraining.
CoRR
Dylan Sam, Ayan Chakrabarti, Afshin Rostamizadeh, Srikumar Ramalingam, Gui Citovsky, Sanjiv Kumar
2025 J jnl
Evaluating Language Model Reasoning about Confidential Information.
CoRR
Dylan Sam, Alexander Robey, Andy Zou, Matt Fredrikson, J. Zico Kolter
2025 J jnl
Finetuning CLIP to Reason about Pairwise Differences.
Trans. Mach. Learn. Res.
Dylan Sam, Devin Willmott, João D. Semedo, J. Zico Kolter
2025 J jnl
Predicting the Performance of Black-box LLMs through Self-Queries.
CoRR
Dylan Sam, Marc Finzi, J. Zico Kolter
2025 J jnl
Safety Pretraining: Toward the Next Generation of Safe AI.
CoRR
Pratyush Maini, Sachin Goyal, Dylan Sam, Alexander Robey, Yash Savani, Yiding Jiang, Andy Zou, Zacharcy C. Lipton, J. Zico Kolter
2024 A conf
Auditing Fairness under Unobserved Confounding.
AISTATS
Yewon Byun, Dylan Sam, Michael Oberst, Zachary C. Lipton, Bryan Wilder
2024 J jnl
Auditing Fairness under Unobserved Confounding.
CoRR
Yewon Byun, Dylan Sam, Michael Oberst, Zachary C. Lipton, Bryan Wilder
2024 J jnl
Bayesian Neural Networks with Domain Knowledge Priors.
CoRR
Dylan Sam, Rattana Pukdee, Daniel P. Jeong, Yewon Byun, J. Zico Kolter
2024 A conf
Computing Low-Entropy Couplings for Large-Support Distributions.
UAI
Samuel Sokota, Dylan Sam, Christian Schröder de Witt, Spencer Compton, Jakob N. Foerster, J. Zico Kolter
2024 J jnl
Computing Low-Entropy Couplings for Large-Support Distributions.
CoRR
Samuel Sokota, Dylan Sam, Christian Schröder de Witt, Spencer Compton, Jakob N. Foerster, J. Zico Kolter
2024 J jnl
Finetuning CLIP to Reason about Pairwise Differences.
CoRR
Dylan Sam, Devin Willmott, João D. Semedo, J. Zico Kolter
2024 A* conf
Understanding prompt engineering may not require rethinking generalization.
ICLR
Victor Akinwande, Yiding Jiang, Dylan Sam, J. Zico Kolter
2023 A* conf
Label Propagation with Weak Supervision.
ICLR
Rattana Pukdee, Dylan Sam, Pradeep Kumar Ravikumar, Nina Balcan
2023 A* conf
Learning with Explanation Constraints.
NeurIPS
Rattana Pukdee, Dylan Sam, J. Zico Kolter, Maria-Florina Balcan, Pradeep Ravikumar
2023 J jnl
Learning with Explanation Constraints.
CoRR
Rattana Pukdee, Dylan Sam, J. Zico Kolter, Maria-Florina Balcan, Pradeep Ravikumar
2023 A* conf
Losses over Labels: Weakly Supervised Learning via Direct Loss Construction.
AAAI
Dylan Sam, J. Zico Kolter
2023 J jnl
Understanding prompt engineering may not require rethinking generalization.
CoRR
Victor Akinwande, Yiding Jiang, Dylan Sam, J. Zico Kolter
2022 J jnl
Improving self-supervised representation learning via sequential adversarial masking.
CoRR
Dylan Sam, Min Bai, Tristan McKinney, Li Erran Li
2022 J jnl
Label Propagation with Weak Supervision.
CoRR
Rattana Pukdee, Dylan Sam, Maria-Florina Balcan, Pradeep Ravikumar
2022 J jnl
Losses over Labels: Weakly Supervised Learning via Direct Loss Construction.
CoRR
Dylan Sam, J. Zico Kolter
2021 A* conf
Adversarial Multi Class Learning under Weak Supervision with Performance Guarantees.
ICML
Alessio Mazzetto, Cyrus Cousins, Dylan Sam, Stephen H. Bach, Eli Upfal
2021 A conf
Semi-Supervised Aggregation of Dependent Weak Supervision Sources With Performance Guarantees.
AISTATS
Alessio Mazzetto, Dylan Sam, Andrew Park, Eli Upfal, Stephen H. Bach