publication.html

<h2>2024</h2>
<ul>
<li><a href="https://arxiv.org/pdf/2409.12822">Language Models Learn to Mislead Humans via RLHF</a>.<br>Jiaxin Wen, Ruiqi Zhong, Akbir Khan, Ethan Perez, Jacob Steinhardt, Minlie Huang, Sam Boman, He He and Shi Feng. <i>arXiv:2409.12822 preprint</i>, 2024.  [<a href="javascript:copy(div0, bib0)">bib</a>]<br>
<div id="div0"></div><div id="bib0" style="display:none">
<div class="bib">
<pre>
@article{wen2024language,
        author={Jiaxin Wen and Ruiqi Zhong and Akbir Khan and Ethan Perez and Jacob Steinhardt and Minlie Huang and Sam Boman and He He and Shi Feng},
        title={Language Models Learn to Mislead Humans via RLHF},
        journal={arXiv:2409.12822},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2407.04549">Spontaneous Reward Hacking in Iterative Self-Refinement</a>.<br>Jane Pan, He He, Sam Bowman and Shi Feng. <i>arXiv:2407.04549 preprint</i>, 2024.  [<a href="javascript:copy(div1, bib1)">bib</a>]<br>
<div id="div1"></div><div id="bib1" style="display:none">
<div class="bib">
<pre>
@article{pan2024spontaneous,
        author={Jane Pan and He He and Sam Bowman and Shi Feng},
        title={Spontaneous Reward Hacking in Iterative Self-Refinement},
        journal={arXiv:2407.04549},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2406.12158">LLMs Are Prone to Fallacies in Causal Inference</a>.<br>Nitish Joshi, Abu Saparov, Yixin Wang and He He. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2024.  [<a href="javascript:copy(div2, bib2)">bib</a>]<br>
<div id="div2"></div><div id="bib2" style="display:none">
<div class="bib">
<pre>
@inproceedings{joshi2024prone,
        author={Nitish Joshi and Abu Saparov and Yixin Wang and He He},
        title={LLMs Are Prone to Fallacies in Causal Inference},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2404.19733">Iterative Reasoning Preference Optimization</a>.<br>Richard Yuanzhe Pang, Weizhe Yuan, Kyunghyun Cho, He He, Sainbayar Sukhbaatar and Jason Weston. <i>Neural Information Processing Systems (NeurIPS)</i>, 2024.  [<a href="javascript:copy(div3, bib3)">bib</a>]<br>
<div id="div3"></div><div id="bib3" style="display:none">
<div class="bib">
<pre>
@inproceedings{pang2024iterative,
        author={Richard Yuanzhe Pang and Weizhe Yuan and Kyunghyun Cho and He He and Sainbayar Sukhbaatar and Jason Weston},
        title={Iterative Reasoning Preference Optimization},
        booktitle={Neural Information Processing Systems (NeurIPS)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2404.16019">The {PRISM} Alignment Project: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models</a>.<br>Hannah Rose Kirk, Alexander Whitefield, Paul Röttger, Andrew Bean, Katerina Margatina, Juan Ciro, Rafael Mosquera, Max Bartolo, Adina Williams, He He, Bertie Vidgen and Scott A Hale. <i>Neural Information Processing Systems (NeurIPS)</i>, 2024.  <font color="red">Oral</font> [<a href="javascript:copy(div4, bib4)">bib</a>]<br>
<div id="div4"></div><div id="bib4" style="display:none">
<div class="bib">
<pre>
@inproceedings{kirk2024prism,
        author={Hannah Rose Kirk and Alexander Whitefield and Paul Röttger and Andrew Bean and Katerina Margatina and Juan Ciro and Rafael Mosquera and Max Bartolo and Adina Williams and He He and Bertie Vidgen and Scott A Hale},
        title={The \{PRISM\} Alignment Project: What Participatory, Representative and Individualised Human Feedback Reveals About the Subjective and Multicultural Alignment of Large Language Models},
        booktitle={Neural Information Processing Systems (NeurIPS)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2404.09932">Foundational Challenges in Assuring Alignment and Safety of Large Language Models</a>.<br>Usman Anwar, Abulhair Saparov, Javier Rando, Daniel Paleka, Miles Turpin, Peter Hase, Ekdeep Singh Lubana, Erik Jenner, Stephen Casper, Oliver Sourbut, Benjamin L Edelman, Zhaowei Zhang, Mario Günther, Anton Korinek, Jose Hernandez-Orallo, Lewis Hammond, Eric Bigelow, Alexander Pan, Lauro Langosco, Tomasz Korbak, Heidi Zhang, Ruiqi Zhong, Sean O hEigeartaigh, Gabriel Recchia, Giulio Corsi, Alan Chan, Markus Anderljung, Lilian Edwards, Yoshua Bengio, Danqi Chen, Samuel Albanie, Tegan Maharaj, Jakob Foerster, Florian Tramer, He He, Atoosa Kasirzadeh, Yejin Choi and David Krueger. <i>Transaction on Machine Learning Research (TMLR)</i>, 2024.  [<a href="javascript:copy(div5, bib5)">bib</a>]<br>
<div id="div5"></div><div id="bib5" style="display:none">
<div class="bib">
<pre>
@article{anwar2024foundational,
        author={Usman Anwar and Abulhair Saparov and Javier Rando and Daniel Paleka and Miles Turpin and Peter Hase and Ekdeep Singh Lubana and Erik Jenner and Stephen Casper and Oliver Sourbut and Benjamin L Edelman and Zhaowei Zhang and Mario Günther and Anton Korinek and Jose Hernandez-Orallo and Lewis Hammond and Eric Bigelow and Alexander Pan and Lauro Langosco and Tomasz Korbak and Heidi Zhang and Ruiqi Zhong and Sean O hEigeartaigh and Gabriel Recchia and Giulio Corsi and Alan Chan and Markus Anderljung and Lilian Edwards and Yoshua Bengio and Danqi Chen and Samuel Albanie and Tegan Maharaj and Jakob Foerster and Florian Tramer and He He and Atoosa Kasirzadeh and Yejin Choi and David Krueger},
        title={Foundational Challenges in Assuring Alignment and Safety of Large Language Models},
        journal={TMLR},
        volume={},
        pages={},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2404.00246">Your Co-Workers Matter: Evaluating Collaborative Capabilities of Language Models in Blocks World</a>.<br>Guande Wu, Chen Zhao, Claudio Silva and He He. <i>Findings of the Association for Computational Linguistics (ACL Findings)</i>, 2024.  [<a href="javascript:copy(div6, bib6)">bib</a>]
[<a href="https://github.com/jnzs1836/coblock">code</a>]<br>
<div id="div6"></div><div id="bib6" style="display:none">
<div class="bib">
<pre>
@inproceedings{wu2024coblock,
        author={Guande Wu and Chen Zhao and Claudio Silva and He He},
        title={Your Co-Workers Matter: Evaluating Collaborative Capabilities of Language Models in Blocks World},
        booktitle={Findings of the Association for Computational Linguistics (ACL Findings)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2402.12530">Parallel Structures in Pre-training Data Yield In-Context Learning</a>.<br>Yanda Chen, Chen Zhao, Zhou Yu, Kathleen McKeown and He He. <i>Association for Computational Linguistics (ACL)</i>, 2024.  [<a href="javascript:copy(div7, bib7)">bib</a>]
[<a href="https://github.com/yandachen/ParallelStructuresICL">code</a>]<br>
<div id="div7"></div><div id="bib7" style="display:none">
<div class="bib">
<pre>
@inproceedings{chen2024parallel,
        author={Yanda Chen and Chen Zhao and Zhou Yu and Kathleen McKeown and He He},
        title={Parallel Structures in Pre-training Data Yield In-Context Learning},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2401.13986">Towards Consistent Natural-Language Explanations via Explanation-Consistency Finetuning</a>.<br>Yanda Chen, Chandan Singh, Xiaodong Liu, Simiao Zuo, Bin Yu, He He and Jianfeng Gao. <i>arXiv:2401.13986 preprint</i>, 2024.  [<a href="javascript:copy(div8, bib8)">bib</a>]
[<a href="https://github.com/yandachen/explanation-consistency-finetuning">code</a>]<br>
<div id="div8"></div><div id="bib8" style="display:none">
<div class="bib">
<pre>
@article{chen2024towards,
        author={Yanda Chen and Chandan Singh and Xiaodong Liu and Simiao Zuo and Bin Yu and He He and Jianfeng Gao},
        title={Towards Consistent Natural-Language Explanations via Explanation-Consistency Finetuning},
        journal={arXiv:2401.13986},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://www.nature.com/articles/s41586-023-06747-5">Solving Olympiad Geometry without Human Demonstrations</a>.<br>Trieu Trinh, Yuhuai Wu, Quoc V Le, He He and Thang Luong. <i>Nature (Nature)</i>, 2024.  [<a href="javascript:copy(div9, bib9)">bib</a>]<br>
<div id="div9"></div><div id="bib9" style="display:none">
<div class="bib">
<pre>
@article{trinh2024geometry,
        author={Trieu Trinh and Yuhuai Wu and Quoc V Le and He He and Thang Luong},
        title={Solving Olympiad Geometry without Human Demonstrations},
        journal={Nature},
        volume={625},
        pages={476--482},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://openreview.net/pdf?id=wxfqhp9bNR">Improving Multi-Hop Reasoning in LLMs by Learning from Rich Human Feedback</a>.<br>Nitish Joshi, Koushik Kalyanaraman, Zhiting Hu, Kumar Chellapilla, He He and Li Erran Li. <i>AAAI Workshop on Neuro-Symbolic Learning and Reasoning in the era of Large Language Models</i>, 2024.  [<a href="javascript:copy(div10, bib10)">bib</a>]<br>
<div id="div10"></div><div id="bib10" style="display:none">
<div class="bib">
<pre>
@inproceedings{joshi2024improving,
        author={Nitish Joshi and Koushik Kalyanaraman and Zhiting Hu and Kumar Chellapilla and He He and Li Erran Li},
        title={Improving Multi-Hop Reasoning in LLMs by Learning from Rich Human Feedback},
        booktitle={AAAI Workshop on Neuro-Symbolic Learning and Reasoning in the era of Large Language Models},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2311.09480">Show Your Work with Confidence: Confidence Bands for Tuning Curves</a>.<br>Nicholas Lourie, Kyunghyun Cho and He He. <i>North American Chapter of the Association for Computational Linguistics (NAACL)</i>, 2024.  [<a href="javascript:copy(div11, bib11)">bib</a>]
[<a href="https: //github.com/nicholaslourie/opda">code</a>]<br>
<div id="div11"></div><div id="bib11" style="display:none">
<div class="bib">
<pre>
@inproceedings{lourie2024show,
        author={Nicholas Lourie and Kyunghyun Cho and He He},
        title={Show Your Work with Confidence: Confidence Bands for Tuning Curves},
        booktitle={North American Chapter of the Association for Computational Linguistics (NAACL)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2310.18168">Personas as a Way to Model Truthfulness in Language Models</a>.<br>Nitish Joshi*, Javier Rando*, Abulhair Saparov, Najoung Kim and He He. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2024.  [<a href="javascript:copy(div12, bib12)">bib</a>]<br>
<div id="div12"></div><div id="bib12" style="display:none">
<div class="bib">
<pre>
@inproceedings{joshi2024persona,
        author={Nitish Joshi and Javier Rando and Abulhair Saparov and Najoung Kim and He He},
        title={Personas as a Way to Model Truthfulness in Language Models},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2309.05196">Does Writing with Language Models Reduce Content Diversity?</a>.<br>Vishakh Padmakumar and He He. <i>International Conference on Learning Representations (ICLR)</i>, 2024.  [<a href="javascript:copy(div13, bib13)">bib</a>]
[<a href="https://github.com/vishakhpk/hai-diversity">code</a>]<br>
<div id="div13"></div><div id="bib13" style="display:none">
<div class="bib">
<pre>
@inproceedings{padmakumar2024writing,
        author={Vishakh Padmakumar and He He},
        title={Does Writing with Language Models Reduce Content Diversity?},
        booktitle={International Conference on Learning Representations (ICLR)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2307.14117">Leveraging Implicit Feedback from Deployment Data in Dialogue</a>.<br>Richard Yuanzhe Pang, Stephen Roller, Kyunghyun Cho, He He and Jason Weston. <i>The European Chapter of the Association for Computational Linguistics (EACL)</i>, 2024.  [<a href="javascript:copy(div14, bib14)">bib</a>]<br>
<div id="div14"></div><div id="bib14" style="display:none">
<div class="bib">
<pre>
@inproceedings{pang2024leveraging,
        author={Richard Yuanzhe Pang and Stephen Roller and Kyunghyun Cho and He He and Jason Weston},
        title={Leveraging Implicit Feedback from Deployment Data in Dialogue},
        booktitle={The European Chapter of the Association for Computational Linguistics (EACL)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2307.08678.pdf">Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanations </a>.<br>Yanda Chen, Ruiqi Zhong, Narutatsu Ri, Chen Zhao, He He, Jacob Steinhardt, Zhou Yu and Kathleen McKeown. <i>International Conference on Machine Learning (ICML)</i>, 2024.  <font color="red">Spotlight</font> [<a href="javascript:copy(div15, bib15)">bib</a>]
[<a href="https://github.com/yandachen/CounterfactualSimulatability">code</a>]<br>
<div id="div15"></div><div id="bib15" style="display:none">
<div class="bib">
<pre>
@inproceedings{chen2024do,
        author={Yanda Chen and Ruiqi Zhong and Narutatsu Ri and Chen Zhao and He He and Jacob Steinhardt and Zhou Yu and Kathleen McKeown},
        title={Do Models Explain Themselves? Counterfactual Simulatability of Natural Language Explanations },
        booktitle={International Conference on Machine Learning (ICML)},
        year={2024}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2210.01302">Nuisances via Negativa: Adjusting for Spurious Correlations via Data Augmentation</a>.<br>Aahlad Puli, Nitish Joshi, He He and Rajesh Ranganath. <i>Transaction on Machine Learning Research (TMLR)</i>, 2024.  [<a href="javascript:copy(div16, bib16)">bib</a>]<br>
<div id="div16"></div><div id="bib16" style="display:none">
<div class="bib">
<pre>
@article{puli2024nuisances,
        author={Aahlad Puli and Nitish Joshi and He He and Rajesh Ranganath},
        title={Nuisances via Negativa: Adjusting for Spurious Correlations via Data Augmentation},
        journal={TMLR},
        volume={},
        pages={},
        year={2024}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2023</h2>
<ul>
<li><a href="https://arxiv.org/abs/2308.06246">{ARGUS}: Visualization of {AI}-Assisted Task Guidance in {AR}</a>.<br>Sonia Castelo, Joao Rulff, Erin McGowan, Bea Steers, Guande Wu, Shaoyu Chen, Iran Roman, Roque Lopez, Ethan Brewer, Chen Zhao, Jing Qian, Kyunghyun Cho, He He, Qi Sun, Huy Vo, Juan Bello, Michael Krone and Claudio Silva. <i>IEEE Transactions on Visualization and Computer Graphics (IEEE Vis)</i>, 2023.  [<a href="javascript:copy(div17, bib17)">bib</a>]<br>
<div id="div17"></div><div id="bib17" style="display:none">
<div class="bib">
<pre>
@inproceedings{castelo2023argus,
        author={Sonia Castelo and Joao Rulff and Erin McGowan and Bea Steers and Guande Wu and Shaoyu Chen and Iran Roman and Roque Lopez and Ethan Brewer and Chen Zhao and Jing Qian and Kyunghyun Cho and He He and Qi Sun and Huy Vo and Juan Bello and Michael Krone and Claudio Silva},
        title={\{ARGUS\}: Visualization of \{AI\}-Assisted Task Guidance in \{AR\}},
        booktitle={IEEE Transactions on Visualization and Computer Graphics (IEEE Vis)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="https://proceedings.mlr.press/v225/nguyen23a/nguyen23a.pdf">Pragmatic Radiology Report Generation</a>.<br>Dang Nguyen, Chacha Chen, He He and Chenhao Tan. <i>Machine Learning for Health (ML4H)</i>, 2023.  [<a href="javascript:copy(div18, bib18)">bib</a>]<br>
<div id="div18"></div><div id="bib18" style="display:none">
<div class="bib">
<pre>
@inproceedings{nguyen2023pragmatic,
        author={Dang Nguyen and Chacha Chen and He He and Chenhao Tan},
        title={Pragmatic Radiology Report Generation},
        booktitle={Machine Learning for Health (ML4H)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2305.15269">Testing the General Deductive Reasoning Capacity of Large Language Models Using OOD Examples</a>.<br>Abulhair Saparov, Richard Yuanzhe Pang, Vishakh Padmakumar, Nitish Joshi, Seyed Mehran Kazemi, Najoung Kim* and He He*. <i>Neural Information Processing Systems (NeurIPS)</i>, 2023.  [<a href="javascript:copy(div19, bib19)">bib</a>]
[<a href="https://github.com/asaparov/prontoqa">code</a>]<br>
<div id="div19"></div><div id="bib19" style="display:none">
<div class="bib">
<pre>
@inproceedings{saparov2023testing,
        author={Abulhair Saparov and Richard Yuanzhe Pang and Vishakh Padmakumar and Nitish Joshi and Seyed Mehran Kazemi and Najoung Kim and He He},
        title={Testing the General Deductive Reasoning Capacity of Large Language Models Using OOD Examples},
        booktitle={Neural Information Processing Systems (NeurIPS)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2305.13299">Measuring Inductive Biases of In-Context Learning with Underspecified Demonstrations</a>.<br>Chenglei Si*, Dan Friedman*, Nitish Joshi, Shi Feng, Danqi Chen and He He. <i>Association for Computational Linguistics (ACL)</i>, 2023.  [<a href="javascript:copy(div20, bib20)">bib</a>]
[<a href="https://github.com/NoviScl/AmbigPrompt">code</a>]<br>
<div id="div20"></div><div id="bib20" style="display:none">
<div class="bib">
<pre>
@inproceedings{si2023measuring,
        author={Chenglei Si and Dan Friedman and Nitish Joshi and Shi Feng and Danqi Chen and He He},
        title={Measuring Inductive Biases of In-Context Learning with Underspecified Demonstrations},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="">Efficient Shapley Values Estimation by Amortization for Text Classification</a>.<br>Chenghao Yang, Fan Yin, He He, Kai-Wei Chang, Xiaofei Ma and Bing Xiang. <i>Association for Computational Linguistics (ACL)</i>, 2023.  [<a href="javascript:copy(div21, bib21)">bib</a>]<br>
<div id="div21"></div><div id="bib21" style="display:none">
<div class="bib">
<pre>
@inproceedings{yang2023efficient,
        author={Chenghao Yang and Fan Yin and He He and Kai-Wei Chang and Xiaofei Ma and Bing Xiang},
        title={Efficient Shapley Values Estimation by Amortization for Text Classification},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2211.08714">Reward Gaming in Conditional Text Generation</a>.<br>Richard Yuanzhe Pang, Vishakh Padmakumar, Thibault Sellam, Ankur P Parikh and He He. <i>Association for Computational Linguistics (ACL)</i>, 2023.  [<a href="javascript:copy(div22, bib22)">bib</a>]
[<a href="{% link /docs/presentation/reward-gaming-slides.pdf %}">talk</a>]<br>
<div id="div22"></div><div id="bib22" style="display:none">
<div class="bib">
<pre>
@inproceedings{pang2023reward,
        author={Richard Yuanzhe Pang and Vishakh Padmakumar and Thibault Sellam and Ankur P Parikh and He He},
        title={Reward Gaming in Conditional Text Generation},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2303.04562.pdf">Extrapolative Controlled Sequence Generation via Iterative Refinement </a>.<br>Vishakh Padmakumar, Richard Yuanzhe Pang, He He and Ankur P Parikh. <i>International Conference on Machine Learning (ICML)</i>, 2023.  [<a href="javascript:copy(div23, bib23)">bib</a>]
[<a href="https://github.com/vishakhpk/iter-extrapolation">code</a>]<br>
<div id="div23"></div><div id="bib23" style="display:none">
<div class="bib">
<pre>
@inproceedings{padmakumar2023extrapolative,
        author={Vishakh Padmakumar and Richard Yuanzhe Pang and He He and Ankur P Parikh},
        title={Extrapolative Controlled Sequence Generation via Iterative Refinement },
        booktitle={International Conference on Machine Learning (ICML)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2210.04782">Robustification of Multilingual Language Models to Real-world Noise in Crosslingual Zero-shot Settings with Robust Contrastive Pretraining</a>.<br>Asa Cooper Stickland*, Sailik Sengupta*, Jason Krone, He He and Saab Mansour. <i>The European Chapter of the Association for Computational Linguistics (EACL)</i>, 2023.  [<a href="javascript:copy(div24, bib24)">bib</a>]
[<a href="https://github.com/amazon-science/multilingual-robust-contrastive-pretraining">code</a>]<br>
<div id="div24"></div><div id="bib24" style="display:none">
<div class="bib">
<pre>
@inproceedings{stickland2023robustification,
        author={Asa Cooper Stickland and Sailik Sengupta and Jason Krone and He He and Saab Mansour},
        title={Robustification of Multilingual Language Models to Real-world Noise in Crosslingual Zero-shot Settings with Robust Contrastive Pretraining},
        booktitle={The European Chapter of the Association for Computational Linguistics (EACL)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="https://aclanthology.org/2023.findings-eacl.70.pdf">How do decoding algorithms distribute information in dialogue responses?</a>.<br>Saranya Venkatraman, He He and David Reitter. <i>Findings of the European Chapter of the Association for Computational Linguistics (EACL Findings)</i>, 2023.  [<a href="javascript:copy(div25, bib25)">bib</a>]<br>
<div id="div25"></div><div id="bib25" style="display:none">
<div class="bib">
<pre>
@inproceedings{venkatraman2023decoding,
        author={Saranya Venkatraman and He He and David Reitter},
        title={How do decoding algorithms distribute information in dialogue responses?},
        booktitle={Findings of the European Chapter of the Association for Computational Linguistics (EACL Findings)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2210.01240">Language Models are Greedy Reasoners: A Systematic Formal Analysis of Chain-of-Thought</a>.<br>Abulhair Saparov and He He. <i>International Conference on Learning Representations (ICLR)</i>, 2023.  [<a href="javascript:copy(div26, bib26)">bib</a>]
[<a href="http://github.com/asaparov/prontoqa">code</a>]
[<a href="{% link /docs/presentation/prontoqa-slides.pdf %}">talk</a>]<br>
<div id="div26"></div><div id="bib26" style="display:none">
<div class="bib">
<pre>
@inproceedings{saparov2023language,
        author={Abulhair Saparov and He He},
        title={Language Models are Greedy Reasoners: A Systematic Formal Analysis of Chain-of-Thought},
        booktitle={International Conference on Learning Representations (ICLR)},
        year={2023}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2209.07661">On the Relation between Sensitivity and Accuracy in In-context Learning</a>.<br>Yanda Chen, Chen Zhao, Zhou Yu, Kathleen McKeown and He He. <i>Findings of the Empirical Methods in Natural Language Processing (EMNLP Findings)</i>, 2023.  [<a href="javascript:copy(div27, bib27)">bib</a>]<br>
<div id="div27"></div><div id="bib27" style="display:none">
<div class="bib">
<pre>
@inproceedings{chen2023relation,
        author={Yanda Chen and Chen Zhao and Zhou Yu and Kathleen McKeown and He He},
        title={On the Relation between Sensitivity and Accuracy in In-context Learning},
        booktitle={Findings of the Empirical Methods in Natural Language Processing (EMNLP Findings)},
        year={2023}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2022</h2>
<ul>
<li><a href="https://arxiv.org/pdf/2210.14011.pdf">Are All Spurious Features in Natural Language Alike? An Analysis through a Causal Lens</a>.<br>Nitish Joshi, Xiang Pan and He He. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2022.  [<a href="javascript:copy(div28, bib28)">bib</a>]
[<a href="https://github.com/joshinh/spurious-correlations-nlp">code</a>]
[<a href="{% link /docs/presentation/spurious-features-slides.pdf %}">talk</a>]<br>
<div id="div28"></div><div id="bib28" style="display:none">
<div class="bib">
<pre>
@inproceedings{joshi2022all,
        author={Nitish Joshi and Xiang Pan and He He},
        title={Are All Spurious Features in Natural Language Alike? An Analysis through a Causal Lens},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2210.13669.pdf">Help me write a poem: Instruction Tuning as a Vehicle for Collaborative Poetry Writing</a>.<br>Tuhin Chakrabarty, Vishakh Padmakumar and He He. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2022.  [<a href="javascript:copy(div29, bib29)">bib</a>]
[<a href="https://github.com/vishakhpk/creative-instructions">code</a>]
[<a href="https://copoet-emnlp.github.io">project</a>]<br>
<div id="div29"></div><div id="bib29" style="display:none">
<div class="bib">
<pre>
@inproceedings{chakrabarty2022help,
        author={Tuhin Chakrabarty and Vishakh Padmakumar and He He},
        title={Help me write a poem: Instruction Tuning as a Vehicle for Collaborative Poetry Writing},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://aclanthology.org/2022.emnlp-main.816.pdf">Improving Faithfulness by Augmenting Negative Summaries from Fake Documents</a>.<br>Tianshu Wang, Faisal Ladhak, Esin Durmus and He He. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2022.  [<a href="javascript:copy(div30, bib30)">bib</a>]
[<a href="https://github.com/COFE2022/CoFE">code</a>]<br>
<div id="div30"></div><div id="bib30" style="display:none">
<div class="bib">
<pre>
@inproceedings{wang2022improving,
        author={Tianshu Wang and Faisal Ladhak and Esin Durmus and He He},
        title={Improving Faithfulness by Augmenting Negative Summaries from Fake Documents},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://openreview.net/pdf?id=ZG5Bi1N4V0U">SeqPATE: Differentially Private Text Generation via Knowledge Distillation</a>.<br>Zhiliang Tian, Yingxiu Zhao, Ziyue Huang, Yu-Xiang Wang, Nevin Zhang and He He. <i>Neural Information Processing Systems (NeurIPS)</i>, 2022.  [<a href="javascript:copy(div31, bib31)">bib</a>]<br>
<div id="div31"></div><div id="bib31" style="display:none">
<div class="bib">
<pre>
@inproceedings{tian2022seqpate,
        author={Zhiliang Tian and Yingxiu Zhao and Ziyue Huang and Yu-Xiang Wang and Nevin Zhang and He He},
        title={SeqPATE: Differentially Private Text Generation via Knowledge Distillation},
        booktitle={Neural Information Processing Systems (NeurIPS)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2112.08670">Amortized Noisy Channel Neural Machine Translation</a>.<br>Richard Yuanzhe Pang, He He and Kyunghyun Cho. <i>International Natural Language Generation Conference (INLG)</i>, 2022.  [<a href="javascript:copy(div32, bib32)">bib</a>]<br>
<div id="div32"></div><div id="bib32" style="display:none">
<div class="bib">
<pre>
@inproceedings{pang2022amortized,
        author={Richard Yuanzhe Pang and He He and Kyunghyun Cho},
        title={Amortized Noisy Channel Neural Machine Translation},
        booktitle={International Natural Language Generation Conference (INLG)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/2112.08608.pdf">{QuALITY}: Question Answering with Long Input Texts, Yes!</a>.<br>Richard Yuanzhe Pang, Alicia Parrish, Nitish Joshi, Nikita Nangia, Jason Phang, Angelica Chen, Vishakh Padmakumar, Johnny Ma, Jana Thompson, He He and Sam Bowman. <i>North American Chapter of the Association for Computational Linguistics (NAACL)</i>, 2022.  [<a href="javascript:copy(div33, bib33)">bib</a>]
[<a href="https://github.com/nyu-mll/quality">code</a>]<br>
<div id="div33"></div><div id="bib33" style="display:none">
<div class="bib">
<pre>
@inproceedings{pang2022quality,
        author={Richard Yuanzhe Pang and Alicia Parrish and Nitish Joshi and Nikita Nangia and Jason Phang and Angelica Chen and Vishakh Padmakumar and Johnny Ma and Jana Thompson and He He and Sam Bowman},
        title={\{QuALITY\}: Question Answering with Long Input Texts, Yes!},
        booktitle={North American Chapter of the Association for Computational Linguistics (NAACL)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2204.11117">Exploring the Role of Task Transferability in Large-Scale Multi-Task Learning</a>.<br>Vishakh Padmakumar, Leonard Lausen, Miguel Ballesteros, Sheng Zha, He He and George Karypis. <i>North American Chapter of the Association for Computational Linguistics (NAACL)</i>, 2022.  [<a href="javascript:copy(div34, bib34)">bib</a>]<br>
<div id="div34"></div><div id="bib34" style="display:none">
<div class="bib">
<pre>
@inproceedings{padmakumar2022exploring,
        author={Vishakh Padmakumar and Leonard Lausen and Miguel Ballesteros and Sheng Zha and He He and George Karypis},
        title={Exploring the Role of Task Transferability in Large-Scale Multi-Task Learning},
        booktitle={North American Chapter of the Association for Computational Linguistics (NAACL)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2111.04193">Machine-in-the-Loop Rewriting for Creative Image Captioning</a>.<br>Vishakh Padmakumar and He He. <i>North American Chapter of the Association for Computational Linguistics (NAACL)</i>, 2022.  [<a href="javascript:copy(div35, bib35)">bib</a>]
[<a href="https://github.com/vishakhpk/mil-creative-captioning">code</a>]<br>
<div id="div35"></div><div id="bib35" style="display:none">
<div class="bib">
<pre>
@inproceedings{padmakumar2022machine,
        author={Vishakh Padmakumar and He He},
        title={Machine-in-the-Loop Rewriting for Creative Image Captioning},
        booktitle={North American Chapter of the Association for Computational Linguistics (NAACL)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2110.07814">Meta-learning via Language Model In-context Tuning</a>.<br>Yanda Chen, Ruiqi Zhong, Sheng Zha, George Karypis and He He. <i>Association for Computational Linguistics (ACL)</i>, 2022.  [<a href="javascript:copy(div36, bib36)">bib</a>]
[<a href="https://github.com/yandachen/In-context-Tuning">code</a>]<br>
<div id="div36"></div><div id="bib36" style="display:none">
<div class="bib">
<pre>
@inproceedings{chen2022meta,
        author={Yanda Chen and Ruiqi Zhong and Sheng Zha and George Karypis and He He},
        title={Meta-learning via Language Model In-context Tuning},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2108.13684">Faithful or Extractive? On Mitigating the Faithfulness-Abstractiveness Trade-off in Abstractive Summarization</a>.<br>Faisal Ladhak, Esin Durmus, He He, Claire Cardie and Kathleen McKeown. <i>Association for Computational Linguistics (ACL)</i>, 2022.  [<a href="javascript:copy(div37, bib37)">bib</a>]<br>
<div id="div37"></div><div id="bib37" style="display:none">
<div class="bib">
<pre>
@inproceedings{ladhak2022faithful,
        author={Faisal Ladhak and Esin Durmus and He He and Claire Cardie and Kathleen McKeown},
        title={Faithful or Extractive? On Mitigating the Faithfulness-Abstractiveness Trade-off in Abstractive Summarization},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2022}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2107.00753">An Investigation of the (In)effectiveness of Counterfactually Augmented Data</a>.<br>Nitish Joshi and He He. <i>Association for Computational Linguistics (ACL)</i>, 2022.  [<a href="javascript:copy(div38, bib38)">bib</a>]
[<a href="https://github.com/joshinh/investigation-cad">code</a>]<br>
<div id="div38"></div><div id="bib38" style="display:none">
<div class="bib">
<pre>
@inproceedings{joshi2022investigation,
        author={Nitish Joshi and He He},
        title={An Investigation of the (In)effectiveness of Counterfactually Augmented Data},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2022}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2021</h2>
<ul>
<li><a href="https://proceedings.neurips.cc/paper/2021/file/972cda1e62b72640cb7ac702714a115f-Paper.pdf">{IRM} - When It Works and When It Doesn't: A Test Case of Natural Language Inference</a>.<br>Yana Dranker, He He and Yonatan Belinkov. <i>Neural Information Processing Systems (NeurIPS)</i>, 2021.  [<a href="javascript:copy(div39, bib39)">bib</a>]
[<a href="https://github.com/technion-cs-nlp/irm-for-nli">code</a>]<br>
<div id="div39"></div><div id="bib39" style="display:none">
<div class="bib">
<pre>
@inproceedings{dranker2021irm,
        author={Yana Dranker and He He and Yonatan Belinkov},
        title={\{IRM\} - When It Works and When It Doesn't: A Test Case of Natural Language Inference},
        booktitle={Neural Information Processing Systems (NeurIPS)},
        year={2021}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2109.06827">Types of Out-of-Distribution Texts and How to Detect Them</a>.<br>Udit Arora, William Huang and He He. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2021.  [<a href="javascript:copy(div40, bib40)">bib</a>]
[<a href="https://github.com/uditarora/ood-text-emnlp">code</a>]<br>
<div id="div40"></div><div id="bib40" style="display:none">
<div class="bib">
<pre>
@inproceedings{arora2021types,
        author={Udit Arora and William Huang and He He},
        title={Types of Out-of-Distribution Texts and How to Detect Them},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2021}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2102.06272">Unsupervised Extractive Summarization with Pointwise Mutual Information</a>.<br>Vishakh Padmakumar and He He. <i>The European Chapter of the Association for Computational Linguistics (EACL)</i>, 2021.  [<a href="javascript:copy(div41, bib41)">bib</a>]
[<a href="https://github.com/vishakhpk/mi-unsup-summ">code</a>]<br>
<div id="div41"></div><div id="bib41" style="display:none">
<div class="bib">
<pre>
@inproceedings{padmakumar2021unsupervised,
        author={Vishakh Padmakumar and He He},
        title={Unsupervised Extractive Summarization with Pointwise Mutual Information},
        booktitle={The European Chapter of the Association for Computational Linguistics (EACL)},
        year={2021}
}
</pre>
</div>
</div> </li>
<li><a href="https://openreview.net/pdf?id=RovX-uQ1Hua">Text Generation by Learning from Demonstrations</a>.<br>Richard Yuanzhe Pang and He He. <i>International Conference on Learning Representations (ICLR)</i>, 2021.  [<a href="javascript:copy(div42, bib42)">bib</a>]
[<a href="https://github.com/yzpang/gold-off-policy-text-gen-iclr21">code</a>]
[<a href="{% link /docs/presentation/gold-slides.pdf %}">talk</a>]<br>
<div id="div42"></div><div id="bib42" style="display:none">
<div class="bib">
<pre>
@inproceedings{pang2021text,
        author={Richard Yuanzhe Pang and He He},
        title={Text Generation by Learning from Demonstrations},
        booktitle={International Conference on Learning Representations (ICLR)},
        year={2021}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2020</h2>
<ul>
<li><a href="https://arxiv.org/abs/2007.06778">An Empirical Study on Robustness to Spurious Correlations using Pre-trained Language Models</a>.<br>Lifu Tu, Garima Lalwani, Spandana Gella and He He. <i>Transaction of Association for Computational Linguistics (TACL)</i>, 2020.  [<a href="javascript:copy(div43, bib43)">bib</a>]
[<a href="https://github.com/lifu-tu/Study-NLP-Robustness">code</a>]<br>
<div id="div43"></div><div id="bib43" style="display:none">
<div class="bib">
<pre>
@article{tu2020empirical,
        author={Lifu Tu and Garima Lalwani and Spandana Gella and He He},
        title={An Empirical Study on Robustness to Spurious Correlations using Pre-trained Language Models},
        journal={TACL},
        volume={8},
        pages={},
        year={2020}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/2005.03754">FEQA: A Question Answering Evaluation Framework for Faithfulness Assessment in Abstractive Summarization</a>.<br>Esin Durmus, He He and Mona Diab. <i>Association for Computational Linguistics (ACL)</i>, 2020.  [<a href="javascript:copy(div44, bib44)">bib</a>]
[<a href="https://github.com/esdurmus/feqa">code</a>]
[<a href="https://slideslive.com/38929353/feqa-a-question-answering-evaluation-framework-for-faithfulness-assessment-in-abstractive-summarization">talk</a>]<br>
<div id="div44"></div><div id="bib44" style="display:none">
<div class="bib">
<pre>
@inproceedings{durmus2020feqa,
        author={Esin Durmus and He He and Mona Diab},
        title={FEQA: A Question Answering Evaluation Framework for Faithfulness Assessment in Abstractive Summarization},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2020}
}
</pre>
</div>
</div> </li>
<li><a href="http://www.jmlr.org/papers/volume21/19-429/19-429.pdf">GluonCV and GluonNLP: Deep Learning in Computer Vision and Natural Language Processing</a>.<br>Jian Guo, He He, Tong He, Leonard Lausen, Mu Li, Haibin Lin, Xingjian Shi, Chenguang Wang, Junyuan Xie, Sheng Zha, Aston Zhang, Hang Zhang, Zhi Zhang, Zhongyue Zhang, Shuai Zheng and Yi Zhu. <i>Journal of Machine Learning Research (JMLR)</i>, 2020.  [<a href="javascript:copy(div45, bib45)">bib</a>]
[<a href="https://gluon-nlp.mxnet.io">project</a>]<br>
<div id="div45"></div><div id="bib45" style="display:none">
<div class="bib">
<pre>
@article{guo2020gluoncv,
        author={Jian Guo and He He and Tong He and Leonard Lausen and Mu Li and Haibin Lin and Xingjian Shi and Chenguang Wang and Junyuan Xie and Sheng Zha and Aston Zhang and Hang Zhang and Zhi Zhang and Zhongyue Zhang and Shuai Zheng and Yi Zhu},
        title={GluonCV and GluonNLP: Deep Learning in Computer Vision and Natural Language Processing},
        journal={JMLR},
        volume={21},
        pages={1--7},
        year={2020}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2019</h2>
<ul>
<li><a href="https://arxiv.org/pdf/1908.10763.pdf">Unlearn Dataset Bias for Natural Language Inference by Fitting the Residual</a>.<br>He He, Sheng Zha and Haohan Wang. <i>EMNLP Workshop on DeepLo</i>, 2019.  [<a href="javascript:copy(div46, bib46)">bib</a>]
[<a href="https://github.com/hhexiy/debiased">code</a>]
[<a href="{% link /docs/presentation/2019_emnlp_deeplo_poster.pdf %}">poster</a>]<br>
<div id="div46"></div><div id="bib46" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2019unlearn,
        author={He He and Sheng Zha and Haohan Wang},
        title={Unlearn Dataset Bias for Natural Language Inference by Fitting the Residual},
        booktitle={EMNLP Workshop on DeepLo},
        year={2019}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/1904.06828.pdf">Pun Generation with Surprise</a>.<br>He He*, Nanyun Peng* and Percy Liang. <i>North American Chapter of the Association for Computational Linguistics (NAACL)</i>, 2019.  [<a href="javascript:copy(div47, bib47)">bib</a>]
[<a href="https://github.com/hhexiy/pungen">code</a>]
[<a href="https://worksheets.codalab.org/worksheets/0x5a7d0fe35b144ad68998d74891a31ed6">codalab</a>]<br>
<div id="div47"></div><div id="bib47" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2019pun,
        author={He He and Nanyun Peng and Percy Liang},
        title={Pun Generation with Surprise},
        booktitle={North American Chapter of the Association for Computational Linguistics (NAACL)},
        year={2019}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/1904.04792">Quizbowl: The Case for Incremental Question Answering</a>.<br>Petro Rodriguez, Shi Feng, Mohit Iyyer, He He and Jordan Boyd-Graber. <i>arXiv:1904.04792 preprint</i>, 2019.  [<a href="javascript:copy(div48, bib48)">bib</a>]<br>
<div id="div48"></div><div id="bib48" style="display:none">
<div class="bib">
<pre>
@article{rodriguez2019quizbowl,
        author={Petro Rodriguez and Shi Feng and Mohit Iyyer and He He and Jordan Boyd-Graber},
        title={Quizbowl: The Case for Incremental Question Answering},
        journal={arXiv:1904.04792},
        year={2019}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/abs/1909.13426">A Dynamic Strategy Coach for Effective Negotiation</a>.<br>Yiheng Zhou, He He, Alan Black and Yulia Tsvetkov. <i>Special Interest Group on Discource and Dialogue (SigDial)</i>, 2019.  [<a href="javascript:copy(div49, bib49)">bib</a>]
[<a href="https://github.com/zhouyiheng11/Negotiation-Coach">code</a>]<br>
<div id="div49"></div><div id="bib49" style="display:none">
<div class="bib">
<pre>
@inproceedings{zhou2019coach,
        author={Yiheng Zhou and He He and Alan Black and Yulia Tsvetkov},
        title={A Dynamic Strategy Coach for Effective Negotiation},
        booktitle={Special Interest Group on Discource and Dialogue (SigDial)},
        year={2019}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2018</h2>
<ul>
<li><a href="https://arxiv.org/abs/1808.09637">Decoupling Strategy and Generation in Negotiation Dialogues</a>.<br>He He, Derek Chen, Anusha Balakrishnan and Percy Liang. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2018.  [<a href="javascript:copy(div50, bib50)">bib</a>]
[<a href="https://stanfordnlp.github.io/cocoa/">project</a>]<br>
<div id="div50"></div><div id="bib50" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2018decouple,
        author={He He and Derek Chen and Anusha Balakrishnan and Percy Liang},
        title={Decoupling Strategy and Generation in Negotiation Dialogues},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2018}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/1808.07036.pdf">QuAC: Question Answering in Context</a>.<br>Eunsol Choi*, He He*, Mohit Iyyer*, Mark Yatskar*, Wen-tau Yih, Yejin Choi, Percy Liang and Luke Zettlemoyer. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2018.  [<a href="javascript:copy(div51, bib51)">bib</a>]
[<a href="http://quac.ai">project</a>]<br>
<div id="div51"></div><div id="bib51" style="display:none">
<div class="bib">
<pre>
@inproceedings{choi2018quac,
        author={Eunsol Choi and He He and Mohit Iyyer and Mark Yatskar and Wen-tau Yih and Yejin Choi and Percy Liang and Luke Zettlemoyer},
        title={QuAC: Question Answering in Context},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2018}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/1805.04623.pdf">Sharp Nearby, Fuzzy Far Away: How Neural Language Models Use Context</a>.<br>Urvashi Khandelwal, He He, Peng Qi and Dan Jurafsky. <i>Association for Computational Linguistics (ACL)</i>, 2018.  [<a href="javascript:copy(div52, bib52)">bib</a>]
[<a href="https://github.com/urvashik/lm-context-analysis">code</a>]<br>
<div id="div52"></div><div id="bib52" style="display:none">
<div class="bib">
<pre>
@inproceedings{khandelwal2018lm,
        author={Urvashi Khandelwal and He He and Peng Qi and Dan Jurafsky},
        title={Sharp Nearby, Fuzzy Far Away: How Neural Language Models Use Context},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2018}
}
</pre>
</div>
</div> </li>
<li><a href="https://arxiv.org/pdf/1804.06437.pdf">Delete, Retrieve, Generate: a Simple Approach to Sentiment and Style Transfer</a>.<br>Juncen Li, Robin Jia, He He and Percy Liang. <i>North American Chapter of the Association for Computational Linguistics (NAACL)</i>, 2018.  [<a href="javascript:copy(div53, bib53)">bib</a>]
[<a href="https://github.com/lijuncen/Sentiment-and-Style-Transfer">code</a>]<br>
<div id="div53"></div><div id="bib53" style="display:none">
<div class="bib">
<pre>
@inproceedings{li2018style,
        author={Juncen Li and Robin Jia and He He and Percy Liang},
        title={Delete, Retrieve, Generate: a Simple Approach to Sentiment and Style Transfer},
        booktitle={North American Chapter of the Association for Computational Linguistics (NAACL)},
        year={2018}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2017</h2>
<ul>
<li><a href="https://arxiv.org/pdf/1704.07130.pdf">Learning Symmetric Collaborative Dialogue Agents with Dynamic Knowledge Graph Embeddings</a>.<br>He He, Anusha Balakrishnan, Mihail Eric and Percy Liang. <i>Association for Computational Linguistics (ACL)</i>, 2017.  [<a href="javascript:copy(div54, bib54)">bib</a>]
[<a href="https://stanfordnlp.github.io/cocoa/">project</a>]<br>
<div id="div54"></div><div id="bib54" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2017symmetric,
        author={He He and Anusha Balakrishnan and Mihail Eric and Percy Liang},
        title={Learning Symmetric Collaborative Dialogue Agents with Dynamic Knowledge Graph Embeddings},
        booktitle={Association for Computational Linguistics (ACL)},
        year={2017}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2016</h2>
<ul>
<li><a href="http://arxiv.org/abs/1406.1837">Credit Assignment Compiler for Joint Prediction</a>.<br>Kai-Wei Chang, He He, Hal Daume III, John Langford and St&eacutephane Ross. <i>Neural Information Processing Systems (NeurIPS)</i>, 2016.  [<a href="javascript:copy(div55, bib55)">bib</a>]
[<a href="https://github.com/JohnLangford/vowpal_wabbit">code</a>]<br>
<div id="div55"></div><div id="bib55" style="display:none">
<div class="bib">
<pre>
@inproceedings{chang2016credit,
        author={Kai-Wei Chang and He He and Hal {Daum\'{e} III} and John Langford and St&eacutephane Ross},
        title={Credit Assignment Compiler for Joint Prediction},
        booktitle={Neural Information Processing Systems (NeurIPS)},
        year={2016}
}
</pre>
</div>
</div> </li>
<li><a href="docs/papers/2016_icml_opponent.pdf">Opponent Modeling in Deep Reinforcement Learning</a>.<br>He He, Jordan Boyd-Graber, Kevin Kwok and Hal Daume III. <i>International Conference on Machine Learning (ICML)</i>, 2016.  [<a href="javascript:copy(div56, bib56)">bib</a>]
[<a href="https://github.com/hhexiy/opponent">code</a>]
[<a href="{% link /data/qb_data.tgz %}">data</a>]<br>
<div id="div56"></div><div id="bib56" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2016opponent,
        author={He He and Jordan Boyd-Graber and Kevin Kwok and Hal {Daum\'{e} III}},
        title={Opponent Modeling in Deep Reinforcement Learning},
        booktitle={International Conference on Machine Learning (ICML)},
        year={2016}
}
</pre>
</div>
</div> </li>
<li><a href="docs/papers/2016_naacl_interpretese.pdf">Interpretese vs. Translationese: The Uniqueness of Human Strategies in Simultaneous Interpretation</a>.<br>He He, Jordan Boyd-Graber and Hal Daume III. <i>North American Chapter of the Association for Computational Linguistics (NAACL)</i>, 2016.  [<a href="javascript:copy(div57, bib57)">bib</a>]
[<a href="https://github.com/hhexiy/interpretese">code</a>]<br>
<div id="div57"></div><div id="bib57" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2016interpretese,
        author={He He and Jordan Boyd-Graber and Hal {Daum\'{e} III}},
        title={Interpretese vs. Translationese: The Uniqueness of Human Strategies in Simultaneous Interpretation},
        booktitle={North American Chapter of the Association for Computational Linguistics (NAACL)},
        year={2016}
}
</pre>
</div>
</div> </li>
<li><a href="wacv_20q.pdf">Object Detection in 20 Questions</a>.<br>Xi Chen, He He and Larry Davis. <i>Winter Conference on Applications of Computer Vision (WACV)</i>, 2016.  [<a href="javascript:copy(div58, bib58)">bib</a>]<br>
<div id="div58"></div><div id="bib58" style="display:none">
<div class="bib">
<pre>
@inproceedings{chen201620q,
        author={Xi Chen and He He and Larry Davis},
        title={Object Detection in 20 Questions},
        booktitle={Winter Conference on Applications of Computer Vision (WACV)},
        year={2016}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2015</h2>
<ul>
<li><a href="http://arxiv.org/abs/1602.02181">Active Information Acquisition</a>.<br>He He, Paul Mineiro and Nikos Karampatziakis. <i>ICML Workshop on Machine Learning From and For Adaptive User Technologies: From Active Learning & Experimentation to Optimization & Personalization</i>, 2015.  [<a href="javascript:copy(div59, bib59)">bib</a>]
[<a href="{% link /docs/presentation/aia-poster-nipsws2015.pdf %}">poster</a>]<br>
<div id="div59"></div><div id="bib59" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2015aia,
        author={He He and Paul Mineiro and Nikos Karampatziakis},
        title={Active Information Acquisition},
        booktitle={ICML Workshop on Machine Learning From and For Adaptive User Technologies: From Active Learning & Experimentation to Optimization & Personalization},
        year={2015}
}
</pre>
</div>
</div> </li>
<li>Interactive Incremental Question Answering.<br>Jordan Boyd-Graber, Mohit Iyyer, He He and Hal Daume III. <i>Neural Information Processing Systems (NeurIPS) demo</i>, 2015.  <font color="red">Outstanding Demonstration Award</font> <br>
 </li>
<li><a href="docs/papers/2015_simtrans_rewrite.pdf">Syntax-based Rewriting for Simultaneous Machine Translation</a>.<br>He He, Alvin Grissom II, John Morgan, Jordan Boyd-Graber and Hal Daume III. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2015.  [<a href="javascript:copy(div61, bib61)">bib</a>]
[<a href="{% link /code/rewriter.tgz %}">code</a>]
[<a href="{% link /docs/presentation/emnlp15_slides.pdf %}">talk</a>]<br>
<div id="div61"></div><div id="bib61" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2015rewrite,
        author={He He and Alvin {Grissom II} and John Morgan and Jordan Boyd-Graber and Hal {Daum\'{e} III}},
        title={Syntax-based Rewriting for Simultaneous Machine Translation},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2015}
}
</pre>
</div>
</div> </li>
<li><a href="http://arxiv.org/abs/1503.05615">Learning to Search for Dependencies</a>.<br>Kai-Wei Chang, He He, Hal Daume III and John Langford. <i>arXiv:1503.05615 preprint</i>, 2015.  [<a href="javascript:copy(div62, bib62)">bib</a>]
[<a href="https://github.com/JohnLangford/vowpal_wabbit/wiki/Tutorial">code</a>]<br>
<div id="div62"></div><div id="bib62" style="display:none">
<div class="bib">
<pre>
@article{chang2015dep,
        author={Kai-Wei Chang and He He and Hal {Daum\'{e} III} and John Langford},
        title={Learning to Search for Dependencies},
        journal={arXiv:1503.05615},
        year={2015}
}
</pre>
</div>
</div> </li>
<li><a href="http://ieeexplore.ieee.org/document/7266612">Crowdsourcing with Multi-Dimensional Trust</a>.<br>Xiangyang Liu, He He and John Baras. <i>International Conference on Information Fusion (Fusion)</i>, 2015.  [<a href="javascript:copy(div63, bib63)">bib</a>]<br>
<div id="div63"></div><div id="bib63" style="display:none">
<div class="bib">
<pre>
@inproceedings{liu2015multitrust,
        author={Xiangyang Liu and He He and John Baras},
        title={Crowdsourcing with Multi-Dimensional Trust},
        booktitle={International Conference on Information Fusion (Fusion)},
        year={2015}
}
</pre>
</div>
</div> </li>
<li><a href="http://ieeexplore.ieee.org/document/7248482/">Trust-Aware Optimal Crowdsourcing With Budget Constraint</a>.<br>Xiangyang Liu, He He and John Baras. <i>International Conference on Communications (ICC)</i>, 2015.  [<a href="javascript:copy(div64, bib64)">bib</a>]<br>
<div id="div64"></div><div id="bib64" style="display:none">
<div class="bib">
<pre>
@inproceedings{liu2015budget,
        author={Xiangyang Liu and He He and John Baras},
        title={Trust-Aware Optimal Crowdsourcing With Budget Constraint},
        booktitle={International Conference on Communications (ICC)},
        year={2015}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2014</h2>
<ul>
<li><a href="docs/papers/paper-slt.pdf">Temporal Supervised Learning for Inferring a Dialog Policy from Example Conversations</a>.<br>Lihong Li, He He and Jason D. Williams. <i>Spoken Lanugage Technology Workshop (SLT)</i>, 2014.  [<a href="javascript:copy(div65, bib65)">bib</a>]<br>
<div id="div65"></div><div id="bib65" style="display:none">
<div class="bib">
<pre>
@inproceedings{li2014temporal,
        author={Lihong Li and He He and Jason D. Williams},
        title={Temporal Supervised Learning for Inferring a Dialog Policy from Example Conversations},
        booktitle={Spoken Lanugage Technology Workshop (SLT)},
        year={2014}
}
</pre>
</div>
</div> </li>
<li><a href="docs/papers/ilp-bb.pdf">Learning to Search in Branch and Bound Algorithms</a>.<br>He He, Hal Daume III and Jason Eisner. <i>Neural Information Processing Systems (NeurIPS)</i>, 2014.  [<a href="javascript:copy(div66, bib66)">bib</a>]
[<a href="https://github.com/hhexiy/scip-dagger">code</a>]
[<a href="{% link /docs/presentation/ilp-bb-poster-nips2014.pdf %}">poster</a>]<br>
<div id="div66"></div><div id="bib66" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2014bb,
        author={He He and Hal {Daum\'{e} III} and Jason Eisner},
        title={Learning to Search in Branch and Bound Algorithms},
        booktitle={Neural Information Processing Systems (NeurIPS)},
        year={2014}
}
</pre>
</div>
</div> </li>
<li><a href="docs/papers/2014_simtrans.pdf">Don't Until the Final Verb Wait: Reinforcement Learning for Simultaneous Machine Translation</a>.<br>Alvin Grissom II, He He, John Morgan, Jordan Boyd-Graber and Hal Daume III. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2014.  [<a href="javascript:copy(div67, bib67)">bib</a>]
[<a href="https://youtu.be/hVoxXO3F468">talk</a>]<br>
<div id="div67"></div><div id="bib67" style="display:none">
<div class="bib">
<pre>
@inproceedings{grissom2014simtrans,
        author={Alvin {Grissom II} and He He and John Morgan and Jordan Boyd-Graber and Hal {Daum\'{e} III}},
        title={Don't Until the Final Verb Wait: Reinforcement Learning for Simultaneous Machine Translation},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2014}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2013</h2>
<ul>
<li><a href="docs/papers/depFeat.pdf">Dynamic Feature Selection for Dependency Parsing</a>.<br>He He, Hal Daume III and Jason Eisner. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2013.  [<a href="javascript:copy(div68, bib68)">bib</a>]
[<a href="{% link /docs/presentation/depFeat_screencast.mp4 %}">talk</a>]<br>
<div id="div68"></div><div id="bib68" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2013dep,
        author={He He and Hal {Daum\'{e} III} and Jason Eisner},
        title={Dynamic Feature Selection for Dependency Parsing},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2013}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2012</h2>
<ul>
<li><a href="docs/papers/dynafea_im.pdf">Imitation Learning by Coaching</a>.<br>He He, Hal Daume III and Jason Eisner. <i>Neural Information Processing Systems (NeurIPS)</i>, 2012.  [<a href="javascript:copy(div69, bib69)">bib</a>]
[<a href="{% link /docs/presentation/dynafea_im_poster.pdf %}">poster</a>]<br>
<div id="div69"></div><div id="bib69" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2012coaching,
        author={He He and Hal {Daum\'{e} III} and Jason Eisner},
        title={Imitation Learning by Coaching},
        booktitle={Neural Information Processing Systems (NeurIPS)},
        year={2012}
}
</pre>
</div>
</div> </li>
<li><a href="docs/papers/qb.pdf">Besting the Quiz Master: Crowdsourcing Incremental Classification Games</a>.<br>Jordan Boyd-Graber, Brianna Satinoff, He He and Hal Daume III. <i>Empirical Methods in Natural Language Processing (EMNLP)</i>, 2012.  [<a href="javascript:copy(div70, bib70)">bib</a>]<br>
<div id="div70"></div><div id="bib70" style="display:none">
<div class="bib">
<pre>
@inproceedings{boyd-graber2012qb,
        author={Jordan Boyd-Graber and Brianna Satinoff and He He and Hal {Daum\'{e} III}},
        title={Besting the Quiz Master: Crowdsourcing Incremental Classification Games},
        booktitle={Empirical Methods in Natural Language Processing (EMNLP)},
        year={2012}
}
</pre>
</div>
</div> </li>
<li><a href="docs/papers/dynafea_ws.pdf">Cost-sensitive dynamic feature selection</a>.<br>He He, Hal Daume III and Jason Eisner. <i>ICML Workshop on Inferning</i>, 2012.  [<a href="javascript:copy(div71, bib71)">bib</a>]
[<a href="{% link /docs/presentation/dynafea_poster.pdf %}">poster</a>]
[<a href="{% link /docs/presentation/dynafea_slides.pdf %}">talk</a>]<br>
<div id="div71"></div><div id="bib71" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2012dynafea,
        author={He He and Hal {Daum\'{e} III} and Jason Eisner},
        title={Cost-sensitive dynamic feature selection},
        booktitle={ICML Workshop on Inferning},
        year={2012}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2011</h2>
<ul>
<li><a href="docs/papers/srgpr.pdf">Single Image Super-resolution using Gaussian Process Regression</a>.<br>He He and Wan-Chi Siu. <i>Computer Vision and Pattern Recognition (CVPR)</i>, 2011.  [<a href="javascript:copy(div72, bib72)">bib</a>]
[<a href="{% link /code/matlab_srgpr.zip %}">code</a>]
[<a href="{% link /docs/presentation/srgpr_slides.pdf %}">talk</a>]<br>
<div id="div72"></div><div id="bib72" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2011superres,
        author={He He and Wan-Chi Siu},
        title={Single Image Super-resolution using Gaussian Process Regression},
        booktitle={Computer Vision and Pattern Recognition (CVPR)},
        year={2011}
}
</pre>
</div>
</div> </li>
</ul>
<h2>2010</h2>
<ul>
<li><a href="docs/papers/rcsvm.pdf">Rare Class classification with SVM</a>.<br>He He and Ali Ghodsi. <i>International Conference on Pattern Recognition (ICPR)</i>, 2010.  [<a href="javascript:copy(div73, bib73)">bib</a>]
[<a href="{% link /code/rcsvm_matlab.zip %}">code</a>]
[<a href="{% link /docs/presentation/rcsvm_poster.pdf %}">poster</a>]<br>
<div id="div73"></div><div id="bib73" style="display:none">
<div class="bib">
<pre>
@inproceedings{he2010rare,
        author={He He and Ali Ghodsi},
        title={Rare Class classification with SVM},
        booktitle={International Conference on Pattern Recognition (ICPR)},
        year={2010}
}
</pre>
</div>
</div> </li>
</ul>