Sadhika Malladi

37 papers A* 17Journal 20

Year	Rank	Type	Title / Venue / Authors
2025	A*	`conf`	Adaptive Data Optimization: Dynamic Sample Selection with Scaling Laws. ICLR Yiding Jiang, Allan Zhou, Zhili Feng, Sadhika Malladi, J. Zico Kolter
2025	J	`jnl`	In Good GRACEs: Principled Teacher Selection for Knowledge Distillation. CoRR Abhishek Panigrahi, Bingbin Liu, Sadhika Malladi, Sham M. Kakade, Surbhi Goel
2025	A*	`conf`	MUSE: Machine Unlearning Six-Way Evaluation for Language Models. ICLR Weijia Shi, Jaechan Lee, Yangsibo Huang, Sadhika Malladi, Jieyu Zhao, Ari Holtzman, Daogao Liu, Luke Zettlemoyer, Noah A. Smith, Chiyuan Zhang
2025	A*	`conf`	Metadata Conditioning Accelerates Language Model Pre-training. ICML Tianyu Gao, Alexander Wettig, Luxi He, Yihe Dong, Sadhika Malladi, Danqi Chen
2025	J	`jnl`	Metadata Conditioning Accelerates Language Model Pre-training. CoRR Tianyu Gao, Alexander Wettig, Luxi He, Yihe Dong, Sadhika Malladi, Danqi Chen
2025	A*	`conf`	Overtrained Language Models Are Harder to Fine-Tune. ICML Jacob Mitchell Springer, Sachin Goyal, Kaiyue Wen, Tanishq Kumar, Xiang Yue, Sadhika Malladi, Graham Neubig, Aditi Raghunathan
2025	J	`jnl`	Overtrained Language Models Are Harder to Fine-Tune. CoRR Jacob Mitchell Springer, Sachin Goyal, Kaiyue Wen, Tanishq Kumar, Xiang Yue, Sadhika Malladi, Graham Neubig, Aditi Raghunathan
2025	A*	`conf`	Progressive distillation induces an implicit curriculum. ICLR Abhishek Panigrahi, Bingbin Liu, Sadhika Malladi, Andrej Risteski, Surbhi Goel
2025	A*	`conf`	Provable unlearning in topic modeling and downstream tasks. ICLR Stanley Wei, Sadhika Malladi, Sanjeev Arora, Amartya Sanyal
2025	J	`jnl`	The Coverage Principle: How Pre-Training Enables Post-Training. CoRR Fan Chen, Audrey Huang, Noah Golowich, Sadhika Malladi, Adam Block, Jordan T. Ash, Akshay Krishnamurthy, Dylan J. Foster
2025	A*	`conf`	Unintentional Unalignment: Likelihood Displacement in Direct Preference Optimization. ICLR Noam Razin, Sadhika Malladi, Adithya Bhaskar, Danqi Chen, Sanjeev Arora, Boris Hanin
2024	J	`jnl`	Adaptive Data Optimization: Dynamic Sample Selection with Scaling Laws. CoRR Yiding Jiang, Allan Zhou, Zhili Feng, Sadhika Malladi, J. Zico Kolter
2024	A*	`conf`	CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs. NeurIPS Zirui Wang, Mengzhou Xia, Luxi He, Howard Chen, Yitao Liu, Richard Zhu, Kaiqu Liang, Xindi Wu, Haotian Liu, Sadhika Malladi, Alexis Chevalier, Sanjeev Arora, Danqi Chen
2024	J	`jnl`	CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs. CoRR Zirui Wang, Mengzhou Xia, Luxi He, Howard Chen, Yitao Liu, Richard Zhu, Kaiqu Liang, Xindi Wu, Haotian Liu, Sadhika Malladi, Alexis Chevalier, Sanjeev Arora, Danqi Chen
2024	A*	`conf`	LESS: Selecting Influential Data for Targeted Instruction Tuning. ICML Mengzhou Xia, Sadhika Malladi, Suchin Gururangan, Sanjeev Arora, Danqi Chen
2024	J	`jnl`	LESS: Selecting Influential Data for Targeted Instruction Tuning. CoRR Mengzhou Xia, Sadhika Malladi, Suchin Gururangan, Sanjeev Arora, Danqi Chen
2024	J	`jnl`	MUSE: Machine Unlearning Six-Way Evaluation for Language Models. CoRR Weijia Shi, Jaechan Lee, Yangsibo Huang, Sadhika Malladi, Jieyu Zhao, Ari Holtzman, Daogao Liu, Luke Zettlemoyer, Noah A. Smith, Chiyuan Zhang
2024	A*	`conf`	Preference Learning Algorithms Do Not Learn Preference Rankings. NeurIPS Angelica Chen, Sadhika Malladi, Lily H. Zhang, Xinyi Chen, Qiuyi (Richard) Zhang, Rajesh Ranganath, Kyunghyun Cho
2024	J	`jnl`	Preference Learning Algorithms Do Not Learn Preference Rankings. CoRR Angelica Chen, Sadhika Malladi, Lily H. Zhang, Xinyi Chen, Qiuyi Zhang, Rajesh Ranganath, Kyunghyun Cho
2024	J	`jnl`	Progressive distillation induces an implicit curriculum. CoRR Abhishek Panigrahi, Bingbin Liu, Sadhika Malladi, Andrej Risteski, Surbhi Goel
2024	J	`jnl`	Provable unlearning in topic modeling and downstream tasks. CoRR Stanley Wei, Sadhika Malladi, Sanjeev Arora, Amartya Sanyal
2024	A*	`conf`	The Marginal Value of Momentum for Small Learning Rate SGD. ICLR Runzhe Wang, Sadhika Malladi, Tianhao Wang, Kaifeng Lyu, Zhiyuan Li
2024	A*	`conf`	Trainable Transformer in Transformer. ICML Abhishek Panigrahi, Sadhika Malladi, Mengzhou Xia, Sanjeev Arora
2024	J	`jnl`	Unintentional Unalignment: Likelihood Displacement in Direct Preference Optimization. CoRR Noam Razin, Sadhika Malladi, Adithya Bhaskar, Danqi Chen, Sanjeev Arora, Boris Hanin
2023	A*	`conf`	A Kernel-Based View of Language Model Fine-Tuning. ICML Sadhika Malladi, Alexander Wettig, Dingli Yu, Danqi Chen, Sanjeev Arora
2023	A*	`conf`	Fine-Tuning Language Models with Just Forward Passes. NeurIPS Sadhika Malladi, Tianyu Gao, Eshaan Nichani, Alex Damian, Jason D. Lee, Danqi Chen, Sanjeev Arora
2023	J	`jnl`	Fine-Tuning Language Models with Just Forward Passes. CoRR Sadhika Malladi, Tianyu Gao, Eshaan Nichani, Alex Damian, Jason D. Lee, Danqi Chen, Sanjeev Arora
2023	J	`jnl`	The Marginal Value of Momentum for Small Learning Rate SGD. CoRR Runzhe Wang, Sadhika Malladi, Tianhao Wang, Kaifeng Lyu, Zhiyuan Li
2023	J	`jnl`	Trainable Transformer in Transformer. CoRR Abhishek Panigrahi, Sadhika Malladi, Mengzhou Xia, Sanjeev Arora
2022	J	`jnl`	A Kernel-Based View of Language Model Fine-Tuning. CoRR Sadhika Malladi, Alexander Wettig, Dingli Yu, Danqi Chen, Sanjeev Arora
2022	A*	`conf`	On the SDEs and Scaling Rules for Adaptive Gradient Algorithms. NeurIPS Sadhika Malladi, Kaifeng Lyu, Abhishek Panigrahi, Sanjeev Arora
2022	J	`jnl`	On the SDEs and Scaling Rules for Adaptive Gradient Algorithms. CoRR Sadhika Malladi, Kaifeng Lyu, Abhishek Panigrahi, Sanjeev Arora
2021	A*	`conf`	A Mathematical Exploration of Why Language Models Help Solve Downstream Tasks. ICLR Nikunj Saunshi, Sadhika Malladi, Sanjeev Arora
2021	A*	`conf`	On the Validity of Modeling SGD with Stochastic Differential Equations (SDEs). NeurIPS Zhiyuan Li, Sadhika Malladi, Sanjeev Arora
2021	J	`jnl`	On the Validity of Modeling SGD with Stochastic Differential Equations (SDEs). CoRR Zhiyuan Li, Sadhika Malladi, Sanjeev Arora
2020	J	`jnl`	A Mathematical Exploration of Why Language Models Help Solve Downstream Tasks. CoRR Nikunj Saunshi, Sadhika Malladi, Sanjeev Arora
2016	J	`jnl`	EMDomics: a robust and powerful method for the identification of genes differentially expressed between heterogeneous classes. Bioinform. Sheida Nabavi, Daniel Schmolze, Mayinuer Maitituoheti, Sadhika Malladi, Andrew H. Beck