We present SODA: the first publicly available, million-scale high-quality social dialogue
dataset. Using SODA, we train COSMO: a generalizable conversation agent outperforming …

Many real-world applications of language models (LMs), such as writing assistance and
code autocomplete, involve human-LM interaction. However, most benchmarks are non …

Large, transformer-based pretrained language models like BERT, GPT, and T5 have
demonstrated a deep understanding of contextual semantics and language syntax. Their …

While large neural-based conversational models have become increasingly proficient
dialogue agents, recent work has highlighted safety issues with these systems. For example …

Implicit knowledge, such as common sense, is key to fluid human conversations. Current
neural response generation (RG) models are trained to generate responses directly …

Maintaining a consistent persona is essential for building a human-like conversational
model. However, the lack of attention to the partner makes the model more egocentric: they …

Human communication relies on common ground (CG), the mutual knowledge and beliefs
shared by participants, to produce coherent and interesting conversations. In this paper, we …

Warning: this paper contains content that may be offensive or upsetting. Numerous natural
language processing models have tried injecting commonsense by using the ConceptNet …

Large Language Models (LLMs) have showcased exceptional capabilities in various
domains, attracting significant interest from both academia and industry. Despite their …

Target-guided response generation enables dialogue systems to smoothly transition a
conversation from a dialogue context toward a target sentence. Such control is useful for …