Search Results - "Mehri, Shikib"

Overview of the Ninth Dialog System Technology Challenge: DSTC9 by Gunasekara, Chulaka, Kim, Seokhwan, D'Haro, Luis Fernando, Rastogi, Abhinav, Chen, Yun-Nung, Eric, Mihail, Hedayatnia, Behnam, Gopalakrishnan, Karthik, Liu, Yang, Huang, Chao-Wei, Hakkani-Tur, Dilek, Li, Jinchao, Zhu, Qi, Luo, Lingxiao, Liden, Lars, Huang, Kaili, Shayandeh, Shahin, Liang, Runze, Peng, Baolin, Zhang, Zheng, Shukla, Swadheen, Huang, Minlie, Gao, Jianfeng, Mehri, Shikib, Feng, Yulan, Gordon, Carla, Alavi, Seyed Hossein, Traum, David, Eskenazi, Maxine, Beirami, Ahmad, Cho, Eunjoon, Crook, Paul A., De, Ankita, Geramifard, Alborz, Kottur, Satwik, Moon, Seungwhan, Poddar, Shivani, Subba, Rajen

“…This paper introduces the Ninth Dialog System Technology Challenge (DSTC-9). This edition of the DSTC focuses on applying end-to-end dialog technologies for…”
Get full text

GenSF: Simultaneous Adaptation of Generative Pre-trained Models and Slot Filling by Mehri, Shikib, Eskenazi, Maxine

“…In transfer learning, it is imperative to achieve strong alignment between a pre-trained model and a downstream task. Prior work has done this by proposing…”
Get full text

Schema-Guided Paradigm for Zero-Shot Dialog by Mehri, Shikib, Eskenazi, Maxine

“…Developing mechanisms that flexibly adapt dialog systems to unseen tasks and domains is a major challenge in dialog research. Neural models implicitly memorize…”
Get full text

Example-Driven Intent Prediction with Observers by Mehri, Shikib, Eric, Mihail

“…A key challenge of dialog systems research is to effectively and efficiently adapt to new domains. A scalable paradigm for adaptation necessitates the…”
Get full text

Unsupervised Evaluation of Interactive Dialog with DialoGPT by Mehri, Shikib, Eskenazi, Maxine

“…It is important to define meaningful and interpretable automatic evaluation metrics for open-domain dialog research. Standard language generation metrics have…”
Get full text

USR: An Unsupervised and Reference Free Evaluation Metric for Dialog Generation by Mehri, Shikib, Eskenazi, Maxine

“…The lack of meaningful automatic evaluation metrics for dialog has impeded open-domain dialog research. Standard language generation metrics have been shown to…”
Get full text

Multi-Granularity Representations of Dialog by Mehri, Shikib, Eskenazi, Maxine

“…Neural models of dialog rely on generalized latent representations of language. This paper introduces a novel training procedure which explicitly learns…”
Get full text

LAD: Language Models as Data for Zero-Shot Dialog by Mehri, Shikib, Altun, Yasemin, Eskenazi, Maxine

“…To facilitate zero-shot generalization in taskoriented dialog, this paper proposes Language Models as Data (LAD). LAD is a paradigm for creating diverse and…”
Get full text

Middle-Out Decoding by Mehri, Shikib, Sigal, Leonid

“…Despite being virtually ubiquitous, sequence-to-sequence models are challenged by their lack of diversity and inability to be externally controlled. In this…”
Get full text

A Comprehensive Assessment of Dialog Evaluation Metrics by Yeh, Yi-Ting, Eskenazi, Maxine, Mehri, Shikib

“…Automatic evaluation metrics are a crucial component of dialog systems research. Standard language evaluation metrics are known to be ineffective for…”
Get full text

Anchored Preference Optimization and Contrastive Revisions: Addressing Underspecification in Alignment by D'Oosterlinck, Karel, Xu, Winnie, Develder, Chris, Demeester, Thomas, Singh, Amanpreet, Potts, Christopher, Kiela, Douwe, Mehri, Shikib

“…Large Language Models (LLMs) are often aligned using contrastive alignment objectives and preference pair datasets. The interaction between model, paired data,…”
Get full text

The DialPort tools by Huynh, Jessica, Mehri, Shikib, Jiao, Cathy, Eskenazi, Maxine

“…The DialPort project http://dialport.org/, funded by the National Science Foundation (NSF), covers a group of tools and services that aim at fulfilling the…”
Get full text

Reasoning Over History: Context Aware Visual Dialog by Shah, Muhammad A, Mehri, Shikib, Srinivasan, Tejas

“…While neural models have been shown to exhibit strong performance on single-turn visual question answering (VQA) tasks, extending VQA to a multi-turn,…”
Get full text

DialoGLUE: A Natural Language Understanding Benchmark for Task-Oriented Dialogue by Mehri, Shikib, Eric, Mihail, Hakkani-Tur, Dilek

“…A long-standing goal of task-oriented dialogue research is the ability to flexibly adapt dialogue models to new domains. To progress research in this…”
Get full text

CESAR: Automatic Induction of Compositional Instructions for Multi-turn Dialogs by Aksu, Taha, Hazarika, Devamanyu, Mehri, Shikib, Kim, Seokhwan, Hakkani-Tür, Dilek, Liu, Yang, Namazifar, Mahdi

“…Instruction-based multitasking has played a critical role in the success of large language models (LLMs) in multi-turn dialog applications. While publicly…”
Get full text

Data-Efficient Alignment of Large Language Models with Human Feedback Through Natural Language by Jin, Di, Mehri, Shikib, Hazarika, Devamanyu, Padmakumar, Aishwarya, Lee, Sungjin, Liu, Yang, Namazifar, Mahdi

“…Learning from human feedback is a prominent technique to align the output of large language models (LLMs) with human expectations. Reinforcement learning from…”
Get full text

Understanding the Effectiveness of Very Large Language Models on Dialog Evaluation by Huynh, Jessica, Jiao, Cathy, Gupta, Prakhar, Mehri, Shikib, Bajaj, Payal, Chaudhary, Vishrav, Eskenazi, Maxine

“…Language models have steadily increased in size over the past few years. They achieve a high level of performance on various natural language processing (NLP)…”
Get full text

CMU GetGoing: An Understandable and Memorable Dialog System for Seniors by Mehri, Shikib, Black, Alan W, Eskenazi, Maxine

“…Voice-based technologies are typically developed for the average user, and thus generally not tailored to the specific needs of any subgroup of the population,…”
Get full text

Structured Fusion Networks for Dialog by Mehri, Shikib, Srinivasan, Tejas, Eskenazi, Maxine

“…Neural dialog models have exhibited strong performance, however their end-to-end nature lacks a representation of the explicit structure of dialog. This…”
Get full text

Interactive Evaluation of Dialog Track at DSTC9 by Mehri, Shikib, Feng, Yulan, Gordon, Carla, Alavi, Seyed Hossein, Traum, David, Eskenazi, Maxine

“…The ultimate goal of dialog research is to develop systems that can be effectively used in interactive settings by real users. To this end, we introduced the…”
Get full text