CSE Colloquium: Addressing Long-Term Dependencies for Language Generation

Abstract: Building accurate language models that capture meaningful long-term dependencies is a core challenge in natural language processing. Towards this end, I will first present our recent work on a calibration-based approach to measure long-term discrepancies between a generative sequence model and the true distribution. Empirically, we show that state-of-the-art language models, including LSTMs and Transformers, are miscalibrated: the entropy rates of their generations drift dramatically upward over time. We provide provable methods to mitigate this phenomenon. We also show how this calibration-based approach can be used to measure the amount of memory used by language models. Second, I will describe a new algorithm for multi-objective reinforcement learning that allows for effective optimization over multiple (potentially conflicting) long-term goals. Our method demonstrates faster learning as well as greater adaptability on several sequential prediction problems, including task-oriented dialog.

Biography: Karthik Narasimhan is an assistant professor in the Computer Science department at Princeton University. His research spans the areas of natural language processing and reinforcement learning, with a focus on building intelligent agents that learn to operate in the world through both experience and existing human knowledge (e.g. text). Karthik received his PhD from MIT in 2017, and spent a year as a visiting research scientist at OpenAI prior to joining Princeton. His work has received a best paper award at EMNLP 2016 and an honorable mention for best paper at EMNLP 2015.


Share this event:

facebook linked in twitter email

Media Contact: Rebecca Passonneau



The School of Electrical Engineering and Computer Science was created in the spring of 2015 to allow greater access to courses offered by both departments for undergraduate and graduate students in exciting collaborative research in fields.

We offer B.S. degrees in electrical engineering, computer science, computer engineering and data science and graduate degrees (master's degrees and Ph.D.'s) in electrical engineering and computer science and engineering. EECS focuses on the convergence of technologies and disciplines to meet today’s industrial demands.

School of Electrical Engineering and Computer Science

The Pennsylvania State University

207 Electrical Engineering West

University Park, PA 16802


Department of Computer Science and Engineering


Department of Electrical Engineering