Dominican Republic

Language-guided robots performing home and office tasks must navigate in and interact with the world. Grounding language instructions against visual observations and actions to take in an environment is an open challenge. We present Embodied BERT (EmBERT), a transformer-based model which can attend to high-dimensional, multi-modal inputs across long temporal horizons for language-conditioned task completion. Additionally, we bridge the gap between successful object-centric navigation models used for non-interactive agents and the language-guided visual task completion benchmark, ALFRED, by introducing object navigation targets for EmBERT training. EmBERT achieves competitive performance on the ALFRED benchmark, and is the first model to use a full, pretrained BERT stack while handling the long-horizon, dense, multi-modal histories of ALFRED.

EMNLP 2021

Embodied BERT: A Transformer Model for Embodied, Language-guided Visual Task Completion

EMNLP 2021 is planned to be a hybrid event in Punta Cana, Dominican Republic, with both on-site and fully virtual participation possible. The experience for on-site participants would closely approximate a normal pre-COVID *ACL conference, with 5-6 thematically organized parallel sessions and live Q/A and interactive discussion immediately after the talks. Presentations by virtual participants will be equitably interleaved with those of on-site participants, projected on the auditorium screens as if on-site, and also followed immediately by live Q/A and interactive discussion at a time during reasonable waking hours for the virtual presenter. For all participants, on-site and virtual, who are unable to attend a session due to either time-zone issues or because they are participating in another session live, talk recordings and slides will be available online at a minimum after the live presentation (and in many cases before as well), and questions may be submitted in advance on session-specific discussion boards and answered live in session with the usual visual aids if desired.

<iframe style="width:700px;height:400px" src="https://online.fliphtml5.com/ebtyf/ceby/"  seamless="seamless" scrolling="no" frameborder="0" allowtransparency="true" allowfullscreen="true" ></iframe>

Please Note: The EMNLP registration system is not currently connected to the underline site as we are still in the process of building out EMNLP 2021. You will receive access instructions from underline the week of November 1st. 

Access is given only to EMNLP upon registration, if you have not registered please do so [here](https://2021.emnlp.org/registration).

Registered attendees will receive access the week of November 1st.

EMNLP 2021 is planned to be a hybrid event in Punta Cana, Dominican Republic, with both on-site and fully virtual participation possible.

workshop paper

Knowledge Grounded Conversation Models (KGCM) are usually based on a selection/retrieval module and a generation module, trained separately or simultaneously, with or without having access to a "gold" knowledge option. With the introduction of large pre-trained generative models, the selection and generation part have become more and more entangled, shifting the focus towards enhancing knowledge incorporation (from multiple sources) instead of trying to pick the best knowledge option. These approaches however depend on knowledge labels and/or a separate dense retriever for their best performance.
In this work we study the unsupervised selection abilities of pre-trained generative models (e.g. BART) and show that by adding a score-and-aggregate module between encoder and decoder, they are capable of learning to pick the proper knowledge through minimising the language modelling loss (i.e. without having access to knowledge labels). Trained as such, our model - K-Mine - shows competitive selection and generation performance against models that benefit from knowledge labels and/or separate dense retriever.

Teach Me What to Say and I Will Learn What to Pick: \\ Unsupervised Knowledge Selection Through Response Generation with Pretrained Generative Models

Anaphora and ellipses are two common phenomena in dialogues. Without resolving referring expressions and information omission, dialogue systems may fail to generate consistent and coherent responses. Traditionally, anaphora is resolved by coreference resolution and ellipses by query rewrite. In this work, we propose a novel joint learning framework of modeling coreference resolution and query rewriting for complex, multi-turn dialogue understanding. Given an ongoing dialogue between a user and a dialogue assistant, for the user query, our joint learning model first predicts coreference links between the query and the dialogue context, and then generates a self-contained rewritten user query. To evaluate our model, we annotate a dialogue based coreference resolution dataset, MuDoCo, with rewritten queries. Results show that the performance of query rewrite can be substantially boosted (+2.3% F1) with the aid of coreference modeling. Furthermore, our joint model outperforms the state-of-the-art coreference resolution model (+2% F1) on this dataset.

CREAD: Combined Resolution of Ellipses and Anaphora in Dialogues

Endowing a task-oriented dialogue system with adaptiveness to user personality can greatly help improve the performance of a dialogue task. However, such a dialogue system can be practically challenging to implement, because it is unclear how user personality influences dialogue task performance. To explore the relationship between user personality and dialogue task performance, we enrolled participants via crowdsourcing to first answer specified personality questionnaires and then chat with a dialogue system to accomplish assigned tasks. A rule-based dialogue system on the prevalent Multi-Domain Wizard-of-Oz (MultiWOZ) task was used. A total of 211 participants' personalities and their 633 dialogues were collected and analyzed. The results revealed that sociable and extroverted people tended to fail the task, whereas neurotic people were more likely to succeed. We extracted features related to user dialogue behaviors and performed further analysis to determine which kind of behavior influences task performance. As a result, we identified that average utterance length and slots per utterance are the key features of dialogue behavior that are highly correlated with both task performance and user personality.

Influence of user personality on dialogue task performance: A case study using a rule-based dialogue system

Named Entity Recognition (NER) and Entity Linking (EL) play an essential role in voice assistant interaction, but are challenging due to the special difficulties associated with spoken user queries. In this paper, we propose a novel architecture that jointly solves the NER and EL tasks by combining them in a joint reranking module. We show that our proposed framework improves NER accuracy by up to 3.13% and EL accuracy by up to 3.6% in F1 score. The features used also lead to better accuracies in other natural language understanding tasks, such as domain classification and semantic parsing.

Noise Robust Named Entity Understanding for Voice Assistants

Code-mixed language plays a crucial role in communication in multilingual societies. Though the recent growth of web users has greatly boosted the use of such mixed languages, the current generation of dialog systems is primarily monolingual. This increase in the usage of code-mixed language has prompted dialog systems in a similar language. We present our work in Code-Mixed Dialog Generation, an unexplored task in code-mixed languages, generating utterances in code-mixed language rather than a single language that is more often just English. We present a new synthetic corpus in code-mix for dialogs, CM-DailyDialog, by converting an existing English-only dialog corpus to a mixed Hindi-English corpus. We then propose a baseline approach where we show the effectiveness of using mBART like multilingual sequence-to-sequence transformers for code-mixed dialog generation. Our best performing dialog models can conduct coherent conversations in Hindi-English mixed language as evaluated by human and automatic metrics setting new benchmarks for the Code-Mixed Dialog Generation task.

Towards Code-Mixed Hinglish Dialogue Generation

Recently neural response generation models have leveraged large pre-trained transformer models and knowledge snippets to generate relevant and informative responses. However, this does not guarantee that generated responses are factually correct. In this paper, we examine factual correctness in knowledgegrounded neural response generation models. We present a human annotation setup to identify three different response types: responses that are factually consistent with respect to the input knowledge, responses that contain hallucinated knowledge, and non-verifiable chitchat style responses. We use this setup to annotate responses generated using different stateof-the-art models, knowledge snippets, and decoding strategies. In addition, to facilitate the development of a factual consistency detector, we automatically create a new corpus called Conv-FEVER that is adapted from the Wizard of Wikipedia dataset and includes factually consistent and inconsistent responses. We demonstrate the benefit of our Conv-FEVER dataset by showing that the models trained on this data perform reasonably well to detect factually inconsistent responses with respect to the provided knowledge through evaluation on our human annotated data. We will release the Conv-FEVER dataset and the human annotated responses.

Rome was built in 1776: A Case Study on Factual Correctness in Knowledge-Grounded Response Generation

Most prior work on task-oriented dialogue systems is restricted to supporting domain APIs. However, users may have requests that are out of the scope of these APIs. This work focuses on identifying such user requests. Existing methods for this task mainly rely on fine-tuning pre-trained models on large annotated data. We propose a novel method, REDE, based on adaptive representation learning and density estimation. REDE can be applied to zero-shot cases, and quickly learns a high-performing detector with only a few shots by updating less than 3K parameters. We demonstrate REDE's competitive performance on DSTC9 data and our newly collected test set.

Towards Zero and Few-shot Knowledge-seeking Turn Detection in Task-orientated Dialogue Systems

An interactive instruction following task has been proposed as a benchmark for learning to map natural language instructions and first-person vision into sequences of actions to interact with objects in a 3D simulated environment.
We find that an existing end-to-end neural model for this task is not robust to variations of objects and language instructions.
We assume that this problem is due to the high sensitiveness of neural feature extraction to small changes in vision and language inputs.
To mitigate this problem, we propose a neuro-symbolic approach that performs reasoning over high-level symbolic representations that are robust to small changes in raw inputs.
Our experiments on the ALFRED dataset show that our approach significantly outperforms the existing model by 18, 52, and 73 points in the success rate on the ToggleObject, PickupObject, and SliceObject subtasks in unseen environments respectively.

Improving the Robustness to Variations of Objects and Instructions with A Neuro-Symbolic Approach for Interactive Instruction Following

Swarm Robotics, which studies collective behaviors of large populations of interacting robots with simple embodied cognition, is an ideal testbed for studying the cultural evolution of language. Our goal is to expand existing simulations in order to explore a dominant theory in evolutionary linguistics—the self-domestication hypothesis—which suggests that increased prosociality led to the sophistication of the cultural niche, enabling the complexification of languages. In our model, robots in multiple nests are engaged in a foraging task (i.e., gathering resources in their environment) while playing a language game. Crucially, we include two novel features: (1) robot individuation: robots have a partner-specific memory, keeping track of the outcomes of past interactions with specific robots; (2) parameterizable prosociality: robots’ tendency to interact is based on experience: successful communication between robots reduces their aggression toward each other and increases their chances of interacting again. First, we show that these manipulations lead to the formation of a classic “in-group bias” where robots favor interaction with some robots over others - a bias which is highly common in social animals in nature but that was so far absent from swarm robotics models. Second, we show that prosociality provides an evolutionary advantage, with higher prosociality values resulting in the collection of more resources. Finally, we show that prosociality modulates the effect of physical distance on lexical convergence, such that low values of innate prosocially lead to more stable sub-swarm divergence, even in relative proximity (i.e., different nests robustly converge on different lexical variants despite being close to each other).

First Steps Toward a Swarm Robotics Model of Self-Domestication and Language Evolution

Machine-in-the-loop writing aims to enable humans to collaborate with models to effectively complete their writing tasks. Prior work in the creative domain has found that providing humans with a machine written draft or sentence level continuations has limited success as the generated text tends to deviate from the human's intentions. We train a rewriting model that, when prompted, modifies targeted spans of text within the user's original draft, enabling the human to retain control over the content while still taking advantage of the strengths of text generation models to introduce descriptive and figurative elements locally in the text. We evaluate the model on it's ability to collaborate with humans on the task of creative image captioning through a user study on Amazon Mechanical Turk. Users report that the model is helpful and third-party evaluation shows that users write more descriptive and figurative captions on average in the collaborative setting compared to a baseline of the human completing the task alone.

Downloads

Next from EMNLP 2021

Teach Me What to Say and I Will Learn What to Pick: \\ Unsupervised Knowledge Selection Through Response Generation with Pretrained Generative Models

Similar lecture

ConRPG: Paraphrase Generation using Contexts as Regularizer