Skip to search formSkip to main contentSkip to account menu

Abstractive Summarization of Reddit Posts with Multi-level Memory Networks

This work collects Reddit TIFU dataset, consisting of 120K posts from the online discussion forum Reddit, and proposes a novel abstractive summarization model named multi-level memory networks (MMN), equipped with multi- level memory to store the information of text from different levels of abstraction.
View on ACL (opens in a new tab)

KLUE: Korean Language Understanding Evaluation

This work introduces Korean Language Understanding Evaluation (KLUE), a collection of 8 Korean natural language understanding (NLU) tasks, including Topic Classification, SemanticTextual Similarity, Natural Language Inference, Named Entity Recognition, Relation Extraction, Dependency Parsing, Machine Reading Comprehension, and Dialogue State Tracking, and provides suitable evaluation metrics and fine-tuning recipes for pretrained language models for each task.
View PDF on arXiv (opens in a new tab)

Perspective-taking and Pragmatics for Generating Empathetic Responses Focused on Emotion Causes

Taking inspiration from social cognition, a generative estimator is used to infer emotion cause words from utterances with no word-level label and a novel method based on pragmatics is introduced to make dialogue models focus on targeted words in the input during generation.
View on ACL (opens in a new tab)

Curiosity-Bottleneck: Exploration By Distilling Task-Specific Novelty

This work shows that Curiosity-Bottleneck learns an effective exploration strategy by robustly measuring the state novelty in distractive environments where state-of-the-art exploration methods often degenerate.
View Paper (opens in a new tab)

ProsocialDialog: A Prosocial Backbone for Conversational Agents

This work introduces ProsocialDialog, the first large-scale multi-turn dialogue dataset to teach conversational agents to respond to problematic content following social norms, and introduces a dialogue safety detection module, Canary, capable of generating RoTs given conversational context, and a socially-informed dialogue agent, Prost.
View on ACL (opens in a new tab)

Will I Sound like Me? Improving Persona Consistency in Dialogues through Pragmatic Self-Consciousness

Inspired by social cognition and pragmatics, existing dialogue agents are endow with public self-consciousness on the fly through an imaginary listener to enforce dialogue agents to refrain from uttering contradiction and improve consistency of existing dialogue models.
View on ACL (opens in a new tab)

SODA: Million-scale Dialogue Distillation with Social Commonsense Contextualization

This work presents SODA: the first publicly available, million-scale high-quality social dialogue dataset, and trains COSMO: a generalizable conversation model that is significantly more natural and consistent on unseen datasets than best-performing conversation models (e.g., GODEL, BlenderBot-1, Koala, Vicuna).
View PDF on arXiv (opens in a new tab)

How Robust are Fact Checking Systems on Colloquial Claims?

It is found that existing fact checking systems that perform well on claims in formal style significantly degenerate on colloquial claims with the same semantics, and it is shown that document retrieval is the weakest spot in the system even vulnerable to filler words, such as “yeah” and “you know”.
View on ACL (opens in a new tab)

Public Self-consciousness for Endowing Dialogue Agents with Consistent Persona

This approach, based on the Rational Speech Acts framework, attempts to maintain consistency in an unsupervised manner requiring neither additional annotations nor pretrained external models to improve consistency in dialogue agents.
View PDF on arXiv (opens in a new tab)
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)