Kianté Brantley — Publications

(He/Him/His)

Assistant Professor
Kempner Institute and School of Engineering and Applied Sciences (SEAS) Harvard University
150 Western Av., Room 6.141, Allston MA 02134

[kdbrantley@g.harvard.edu] [CV]
You can view my full list of publications at [Google Scholar or Semantic Scholar].

Publications:

2025

Scaling Offline RL via Efficient and Expressive Shortcut Models
Espinosa-Dice, Nicolas and Zhang, Yiyi and Chen, Yiding and Guo, Bradley and Oertell, Owen and Swamy, Gokul and Brantley, Kianté and Sun, Wen
Preprint, 2025
[Arxiv]
Accelerating RL for LLM Reasoning with Optimal Advantage Regression
Brantley, Kianté and Chen, Mingyu and Gao, Zhaolin and Lee, Jason D and Sun, Wen and Zhan, Wenhao and Zhang, Xuezhou
Preprint, 2025
[Arxiv]
Value-Guided Search for Efficient Chain-of-Thought Reasoning
Wang, Kaiwen and Zhou, Jin Peng and Chang, Jonathan and Gao, Zhaolin and Kallus, Nathan and Brantley, Kianté and Sun, Wen
Preprint, 2025
[Arxiv]
Q#: Provably Optimal Distributional RL for LLM Post-Training
Zhou, Jin Peng and Wang, Kaiwen and Chang, Jonathan and Gao, Zhaolin and Kallus, Nathan and Weinberger, Kilian Q and Brantley, Kianté and Sun, Wen
Preprint, 2025
[Arxiv]

2024

Diffusing States and Matching Scores: A New Framework for Imitation Learning
Wu, Runzhe and Chen, Yiding and Swamy, Gokul and Brantley, Kianté and Sun, Wen
The International Conference on Learning Representations (ICLR), 2024
[Arxiv]
LLMs Are In-Context Reinforcement Learners
Monea, Giovanni and Bosselut, Antoine and Brantley, Kianté and Artzi, Yoav
Conference On Language Modeling, 2024
[Arxiv]
Regressing the Relative Future: Efficient Policy Optimization for Multi-turn RLHF
Gao, Zhaolin and Zhan, Wenhao and Chang, Jonathan D and Swamy, Gokul and Brantley, Kianté and Lee, Jason D and Sun, Wen
Proceedings of the International Conference on Machine Learning (ICML), 2024
[Arxiv]
Reviewer2: Optimizing Review Generation Through Prompt Generation
Gao, Zhaolin and Brantley, Kianté and Joachims, Thorsten
Preprint, 2024
[Arxiv]
RL for Consistency Models: Faster Reward Guided Text-to-Image Generation
Oertell, Owen and Chang, Jonathan D and Zhang, Yiyi and Brantley, Kianté and Sun, Wen
Reinforcement Learning Conference, 2024
[Arxiv]
REBEL: Reinforcement Learning via Regressing Relative Rewards
Gao, Zhaolin and Chang, Jonathan D and Zhan, Wenhao and Oertell, Owen and Swamy, Gokul and Brantley, Kianté and Joachims, Thorsten and Bagnell, J Andrew and Lee, Jason D and Sun, Wen
Advances in Neural Information Processing Systems (NeurIPS), 2024
[Arxiv]
A Surprising Failure? Multimodal LLMs and the NLVR Challenge
Wu, Anne and Brantley, Kianté and Artzi, Yoav
Preprint, 2024
[Arxiv]
Dataset Reset Policy Optimization for RLHF
Chang, Jonathan D and Shan, Wenhao and Oertell, Owen and Brantley, Kianté and Misra, Dipendra and Lee, Jason D and Sun, Wen
Preprint, 2024
[Arxiv]
When is Transfer Learning Possible?
Phan, My and Brantley, Kianté and Milani, Stephanie and Mehri, Soroush and Swamy, Gokul and Gordon, Geoffrey J
Proceedings of the International Conference on Machine Learning (ICML), 2024
[Arxiv]
Ranking with Long-Term Constraints
Brantley, Kianté and Fang, Zhichong and Dean, Sarah and Joachims, Thorsten
Proceedings of the ACM International Conference on Web Search and Data Mining (WSDM), 2024
[Arxiv]
Coactive Learning for Large Language Models using Implicit User Feedback
Tucker, Aaron David and Brantley, Kianté and Cahall, Adam and Joachims, Thorsten
Proceedings of the International Conference on Machine Learning (ICML), 2024
[Arxiv]
Adversarial Imitation Learning via Boosting
Chang, Jonathan Daniel and Sreenivas, Dhruv and Huang, Yingbing and Brantley, Kianté and Sun, Wen
The International Conference on Learning Representations (ICLR), 2024
[Arxiv]

2023

Interactive Text Generation
Faltings, Felix and Galley, Michel and Brantley, Kianté and Peng, Baolin and Cai, Weixin and Zhang, Yizhe and Gao, Jianfeng and Dolan, William Bill
Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP), 2023
[Arxiv]
Is Reinforcement Learning (Not) for Natural Language Processing: Benchmarks, Baselines, and Building Blocks for Natural Language Policy Optimization
Ramamurthy, Rajkumar and Ammanabrolu, Prithviraj and Brantley, Kianté and Hessel, Jack and Sifa, Rafet and Bauckhage, Christian and Hajishirzi, Hannaneh and Choi, Yejin
The International Conference on Learning Representations (ICLR), 2023
-- Spotlight --
[Arxiv]
Policy-Gradient Training of Language Models for Ranking
Gao, Ge and Chang, Jonathan Daniel and Cardie, Claire and Brantley, Kianté and Joachims, Thorsten
NeurIPS 2023 Workshop Foundation Models for Decision Making Workshop, 2023
[Arxiv]
Learning to Generate Better Than Your LLM
Chang, Jonathan and Brantley, Kianté and Ramamurthy, Rajkumar, and Misra, Dipendra, and Sun, Wen
NeurIPS 2023 Workshop on Instruction Tuning and Instruction Following, 2023
[Arxiv]
lilGym: Natural Language Visual Reasoning with Reinforcement Learning
Wu, Anne and Brantley, Kianté and Kojima, Noriyuki and Artzi, Yoav
Proceedings of the Conference of the Association for Computational Linguistics (ACL), 2023
[Arxiv]

2021

Expert-in-the-Loop for Sequential Decisions and Predictions
Brantley, Kianté
undefined, 2021
[Arxiv]
Successor feature sets: Generalizing successor representations across policies
Brantley, Kianté and Mehri, Soroush and Gordon, Geoff J
Proceedings of the National Conference on Artificial Intelligence (AAAI), 2021
[Arxiv]

2020

Active Imitation Learning with Noisy Guidance
Brantley, Kianté and Sharaf, Amr and Daumé III, Hal
Proceedings of the Conference of the Association for Computational Linguistics (ACL), 2020
[Arxiv]
Constrained episodic reinforcement learning in concave-convex and knapsack settings
Brantley, Kianté and Dudik, Miro and Lykouris, Thodoris and Miryoosefi, Sobhan and Simchowitz, Max and Slivkins, Aleksandrs and Sun, Wen
Advances in Neural Information Processing Systems (NeurIPS), 2020
[Arxiv]

2019

Disagreement-regularized imitation learning
Brantley, Kianté and Sun, Wen and Henaff, Mikael
The International Conference on Learning Representations (ICLR), 2019
-- Spotlight --
[Arxiv]
Non-monotonic sequential text generation
Welleck, Sean and Brantley, Kianté and Iii, Hal Daumé and Cho, Kyunghyun
Proceedings of the International Conference on Machine Learning (ICML), 2019
[Arxiv]
Reinforcement learning with convex constraints
Miryoosefi, Sobhan and Brantley, Kianté and Daumé III, Hal and Dudik, Miro and Schapire, Robert E
Advances in Neural Information Processing Systems (NeurIPS), 2019
[Arxiv]

2017

The UMD Neural Machine Translation Systems at WMT17 Bandit Learning Task
Sharaf, Amr and Feng, Shi and Nguyen, Khanh and Brantley, Kianté and Daumé III, Hal
Proceedings of the Second Conference on Machine Translation, 2017
[Arxiv]

2016

BCAP: An Artificial Neural Network Pruning Technique to Reduce Overfitting
Brantley, Kianté
undefined, 2016
[Arxiv]

2015

Ldaexplore: Visualizing topic models generated using latent dirichlet allocation
Ganesan, Ashwinkumar and Brantley, Kianté and Pan, Shimei and Chen, Jian
extvis Workshop - Intelligent User Interfaces (IUI), 2015
[Arxiv]