---
title: "Pang Wei Koh — all works — Elamur Research"
description: "All 52 works of Pang Wei Koh in the Elamur Research library, the most cited first: our summaries where we have them, else the original."
canonical: "https://research.elamur.ai/free/en/authors/pang-wei-koh/works/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/pang-wei-koh/works/)

[home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / [Pang Wei Koh](https://md.research.elamur.ai/free/en/authors/pang-wei-koh/) / all works

# Pang Wei Koh — all works

52 works · [← Pang Wei Koh](https://md.research.elamur.ai/free/en/authors/pang-wei-koh/)

1. **[Mobility network models of COVID-19 explain inequities and inform reopening ↗](https://doi.org/10.1038/s41586-020-2923-3) Nature · no summary yet**Serina Y. Chang et al.*2020 · 1,532 citations*
2. [**Understanding Black-box Predictions via Influence Functions arXiv:1703.04730**Pang Wei Koh, Percy Liang*2017 · 1,514 citations*](https://md.research.elamur.ai/free/en/papers/koh-2017-understanding-black-box-predictions-influence/)
3. **[FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation](https://md.research.elamur.ai/free/en/papers/min-2023-factscore-fine-grained-atomic-evaluation/) [Conference on Empirical Methods in Natural Language Processing (EMNLP)](https://md.research.elamur.ai/free/en/venues/emnlp/)**Sewon Min, Kalpesh Krishna, Xinxi Lyu, et al.*2023 · 987 citations*
4. **[Distributionally Robust Neural Networks for Group Shifts: On the Importance of Regularization for Worst-Case Generalization ↗](https://arxiv.org/abs/1911.08731) arXiv (Cornell University) · no summary yet**Shiori Sagawa et al.*2019 · 430 citations*
5. **[WILDS: A Benchmark of in-the-Wild Distribution Shifts ↗](https://arxiv.org/abs/2012.07421) CaltechAUTHORS (California Institute of Technology) · no summary yet**Pang Wei Koh et al.*2020 · 405 citations*
6. **[Toward Trustworthy AI Development: Mechanisms for Supporting Verifiable Claims ↗](https://arxiv.org/abs/2004.07213) arXiv:2004.07213 · no summary yet**Miles Brundage et al.*2020 · 320 citations*
7. **[Tiled convolutional neural networks ↗](https://openalex.org/W2147860648) neural information processing systems · no summary yet**Jiquan Ngiam et al.*2010 · 298 citations*
8. **[2 OLMo 2 Furious ↗](https://arxiv.org/abs/2501.00656) arXiv:2501.00656 · no summary yet**Team OLMo et al.*2025 · 262 citations*
9. **[DataComp: In search of the next generation of multimodal datasets ↗](https://arxiv.org/abs/2304.14108) arXiv:2304.14108 · no summary yet**Samir Yitzhak Gadre et al.*2023 · 224 citations*
10. **[Concept Bottleneck Models ↗](https://arxiv.org/abs/2007.04612) arXiv (Cornell University) · no summary yet**Pang Wei Koh et al.*2020 · 206 citations*
11. **[Stronger data poisoning attacks break data sanitization defenses ↗](https://doi.org/10.1007/s10994-021-06119-y) Machine Learning · no summary yet**Pang Wei Koh, Jacob Steinhardt, Percy Liang*2018 · 200 citations*
12. **[Distributionally Robust Neural Networks ↗](https://openalex.org/W2995793065) International Conference on Learning Representations · no summary yet**Shiori Sagawa et al.*2020 · 161 citations*
13. **[Are aligned neural networks adversarially aligned? ↗](https://arxiv.org/abs/2306.15447) arXiv:2306.15447 · no summary yet**Nicholas Carlini et al.*2023 · 157 citations*
14. **[An Investigation of Why Overparameterization Exacerbates Spurious Correlations ↗](https://arxiv.org/abs/2005.04345) arXiv (Cornell University) · no summary yet**Shiori Sagawa et al.*2020 · 147 citations*
15. **[Just Train Twice: Improving Group Robustness without Training Group Information ↗](https://arxiv.org/abs/2107.09044) arXiv (Cornell University) · no summary yet**Evan Zheran Liu et al.*2021 · 125 citations*
16. **[Certified Defenses for Data Poisoning Attacks ↗](https://arxiv.org/abs/1706.03691) arXiv (Cornell University) · no summary yet**Jacob Steinhardt, Pang Wei Koh, Percy Liang*2017 · 103 citations*
17. **[On the Accuracy of Influence Functions for Measuring Group Effects ↗](https://arxiv.org/abs/1905.13289) arXiv (Cornell University) · no summary yet**Pang Wei Koh et al.*2019 · 95 citations*
18. **[MediQ: Question-Asking LLMs and a Benchmark for Reliable Interactive Clinical Reasoning](https://md.research.elamur.ai/free/en/papers/li-2024-mediq-question-asking-llms-benchmark/) [Conference on Neural Information Processing Systems (NeurIPS)](https://md.research.elamur.ai/free/en/venues/neural-information-processing-systems/)**Shuyue Stella Li, Vidhisha Balachandran, Shangbin Feng, et al.*2024 · 74 citations*
19. **[OpenScholar: Synthesizing Scientific Literature with Retrieval-augmented LMs ↗](https://arxiv.org/abs/2411.14199) arXiv:2411.14199 · no summary yet**Akari Asai et al.*2024 · 62 citations*
20. **[The Generative AI Paradox: "What It Can Create, It May Not Understand" ↗](https://arxiv.org/abs/2311.00059) arXiv:2311.00059 · no summary yet**Peter West et al.*2023 · 62 citations*
21. **[Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalization ↗](https://arxiv.org/abs/2107.04649) arXiv (Cornell University) · no summary yet**John J. Miller et al.*2021 · 59 citations*
22. **[Spurious Rewards: Rethinking Training Signals in RLVR ↗](https://arxiv.org/abs/2506.10947) arXiv:2506.10947 · no summary yet**Rulin Shao et al.*2025 · 53 citations*
23. **[ExpBERT: Representation Engineering with Natural Language Explanations ↗](https://aclanthology.org/2020.acl-main.190/) Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · no summary yet**Shikhar Murty, Pang Wei Koh, Percy Liang*2020 · 52 citations*
24. [**Reliable, Adaptable, and Attributable Language Models with Retrieval arXiv:2403.03187**Akari Asai, Zexuan Zhong, Danqi Chen, et al.*2024 · 45 citations*](https://md.research.elamur.ai/free/en/papers/asai-2024-reliable-adaptable-attributable-language-models/)
25. **[Synthesizing scientific literature with retrieval-augmented language models ↗](https://doi.org/10.1038/s41586-025-10072-4) Nature · no summary yet**Akari Asai et al.*2026 · 42 citations*
26. **[ReasonIR: Training Retrievers for Reasoning Tasks ↗](https://arxiv.org/abs/2504.20595) arXiv (Cornell University) · no summary yet**Rulin Shao et al.*2025 · 41 citations*
27. **[Supporting COVID-19 Policy Response with Large-scale Mobility-based Modeling ↗](https://doi.org/10.1145/3447548.3467182) no summary yet**Serina Y. Chang et al.*2021 · 36 citations*
28. **[Scaling Retrieval-Based Language Models with a Trillion-Token Datastore ↗](https://doi.org/10.52202/079017-2896) neural information processing systems · no summary yet**Rulin Shao et al.*2024 · 30 citations*
29. **[Temporal FiLM: Capturing Long-Range Sequence Dependencies with Feature-Wise Modulations ↗](https://arxiv.org/abs/1909.06628) arXiv (Cornell University) · no summary yet**Sawyer Birnbaum et al.*2019 · 21 citations*
30. **[CopyBench: Measuring Literal and Non-Literal Reproduction of Copyright-Protected Text in Language Model Generation ↗](https://aclanthology.org/2024.emnlp-main.844/) Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing · no summary yet**Tong Chen et al.*2024 · 19 citations*
31. **[Using Large Language Models to Promote Health Equity ↗](https://doi.org/10.1056/aip2400889) NEJM AI · no summary yet**Emma Pierson et al.*2023 · 18 citations*
32. **[Extending the WILDS Benchmark for Unsupervised Adaptation ↗](https://arxiv.org/abs/2112.05090) arXiv (Cornell University) · no summary yet**Shiori Sagawa et al.*2021 · 18 citations*
33. **[Multilingual Diversity Improves Vision-Language Representations ↗](https://doi.org/10.52202/079017-2902) neural information processing systems · no summary yet**Thao Nguyen et al.*2024 · 16 citations*
34. **[Selective Classification Can Magnify Disparities Across Groups ↗](https://arxiv.org/abs/2010.14134) arXiv (Cornell University) · no summary yet**Erik Jones et al.*2020 · 15 citations*
35. **[The Delta Learning Hypothesis: Preference Tuning on Weak Data can Yield Strong Gains ↗](https://arxiv.org/abs/2507.06187) arXiv (Cornell University) · no summary yet**Scott Geng et al.*2025 · 10 citations*
36. **[The Unmet Promise of Synthetic Training Images: Using Retrieved Real Images Performs Better ↗](https://doi.org/10.52202/079017-0254) neural information processing systems · no summary yet**Scott Geng et al.*2024 · 9 citations*
37. **[Inferring Multidimensional Rates of Aging from Cross-Sectional Data ↗](https://arxiv.org/abs/1807.04709) PubMed · no summary yet**Emma Pierson et al.*2018 · 7 citations*
38. **[FlexOLMo: Open Language Models for Flexible Data Use ↗](https://doi.org/10.52202/085713-5530) neural information processing systems · no summary yet**Weijia Shi et al.*2025 · 5 citations*
39. **[Precise Information Control in Long-Form Text Generation ↗](https://doi.org/10.52202/085713-3103) neural information processing systems · no summary yet**Jacqueline He et al.*2025 · 3 citations*
40. **[JobBench: Aligning Agent Work With Human Will ↗](https://arxiv.org/abs/2605.26329) arXiv (Cornell University) · no summary yet**Yuetai Li et al.*2026 · 1 citation*
41. **[ParaPO: Aligning Language Models to Reduce Verbatim Reproduction of Pre-training Data ↗](https://arxiv.org/abs/2504.14452) arXiv (Cornell University) · no summary yet**Tong Chen et al.*2025 · 1 citation*
42. **[Out-of-Domain Robustness via Targeted Augmentations ↗](https://arxiv.org/abs/2302.11861) arXiv (Cornell University) · no summary yet**Irena Gao et al.*2023 · 1 citation*
43. **[Time Series Super Resolution withTemporal Adaptive Batch Normalization ↗](https://openalex.org/W2921699175) no summary yet**Volodymyr Kuleshov et al.*2018 · 1 citation*
44. **[Anchored Decoding: Provably Reducing Copyright Risk for Any Language Model ↗](https://arxiv.org/abs/2602.07120) arXiv (Cornell University) · no summary yet**Jacqueline He et al.*2026*
45. **[Knowledge Distillation During Mid-Training Favors Reasoning over Factual Recall ↗](https://arxiv.org/abs/2609.01532) arXiv:2609.01532 · no summary yet**Jacqueline He et al.*2026*
46. **[ScholarCatalyst: A Benchmark for Retrieving Papers That Inspire New Research ↗](https://arxiv.org/abs/2610.02202) arXiv (Cornell University) · no summary yet**Sohyeon Kim et al.*2026*
47. **[Spokes: Optimizing for Diverse Pretraining Data Selection ↗](https://arxiv.org/abs/2606.15216) arXiv (Cornell University) · no summary yet**Clarence Lee et al.*2026*
48. **[Context Language Models ↗](https://arxiv.org/abs/2609.37725) arXiv (Cornell University) · no summary yet**Rulin Shao et al.*2026*
49. **[Efficient Test-Time Adaptation through Human-AI Interaction ↗](https://arxiv.org/abs/2609.04141) arXiv (Cornell University) · no summary yet**Zora Zhiruo Wang et al.*2026*
50. **[Self-Improving VLM Judges Without Human Annotations ↗](https://arxiv.org/abs/2512.05145) arXiv (Cornell University) · no summary yet**Inna Wanyin Lin et al.*2025*
51. **[Negative Token Merging: Image-based Adversarial Feature Guidance ↗](https://arxiv.org/abs/2412.01339) arXiv (Cornell University) · no summary yet**Jaskirat Singh et al.*2024*
52. **[Overparameterization hurts worst-group accuracy with spurious correlations ↗](https://openalex.org/W3035466577) International Conference on Machine Learning · no summary yet**Shiori Sagawa et al.*2020*