Philip R. Cohen

Levon Haroutunian, Zhuang Li, Lucian Galescu et al.

Large language models (LLMs) have demonstrated impressive capabilities in natural language generation. However, their output quality can be inconsistent, posing challenges for generating natural language from logical forms (LFs). This task requires the generated outputs to embody the exact semantics of LFs, without missing any LF semantics or creating any hallucinations. In this work, we tackle this issue by proposing a novel generate-and-rerank approach. Our approach involves initially generating a set of candidate outputs by prompting an LLM and subsequently reranking them using a task-specific reranker model. In addition, we curate a manually collected dataset to evaluate the alignment between different ranking metrics and human judgements. The chosen ranking metrics are utilized to enhance the training and evaluation of the reranker model. By conducting extensive experiments on three diverse datasets, we demonstrate that the candidates selected by our reranker outperform those selected by baseline methods in terms of semantic consistency and fluency, as measured by three comprehensive metrics. Our findings provide strong evidence for the effectiveness of our approach in improving the quality of generated outputs.

Rachel Love, Edith Law, Philip R. Cohen et al.

Conversational teachable agents offer a promising platform to support learning, both in the classroom and in remote settings. In this context, the agent takes the role of the novice, while the student takes on the role of teacher. This framing is significant for its ability to elicit the Protégé effect in the student-teacher, a pedagogical phenomenon known to increase engagement in the teaching task, and also improve cognitive outcomes. In prior work, teachable agents often take a passive role in the learning interaction, and there are few studies in which the agent and student engage in natural language dialogue during the teaching task. This work investigates the effect of teaching modality when interacting with a virtual agent, via the web-based teaching platform, the Curiosity Notebook. A method of teaching the agent by selecting sentences from source material is compared to a method paraphrasing the source material and typing text input to teach. A user study has been conducted to measure the effect teaching modality on the learning outcomes and engagement of the participants. The results indicate that teaching via paraphrasing and text input has a positive effect on learning outcomes for the material covered, and also on aspects of affective engagement. Furthermore, increased paraphrasing effort, as measured by the similarity between the source material and the material the teacher conveyed to the robot, improves learning outcomes for participants.

Philip R. Cohen, Lucian Galescu, Maayan Shvo

Eva is a neuro-symbolic domain-independent multimodal collaborative dialogue system that takes seriously that the purpose of task-oriented dialogue is to assist the user. To do this, the system collaborates by inferring their intentions and plans, detects obstacles to success, finds plans to overcome them or to achieve higher-level goals, and plans its actions, including speech acts, to help users accomplish those goals. In doing so, the system maintains and reasons with its own declaratively-specified beliefs, goals and intentions, and explicitly reasons about those of its user. Because Eva can track different users' mental states, it can engage multiple agents in multi-party dialogues. Reasoning is accomplished with a modal Horn-clause meta-interpreter that enables computable inference within the subset of logic implemented. The system employs both hierarchical and backward-chaining planning, operating over a rich modal logic-based knowledge and action representation. The planning and reasoning subsystems obey the principles of persistent goals and intentions including: 1) The formation and decomposition of intentions to perform complex actions, 2) the conditions under which persistent goals and intentions can be given up, and 3) persistent goal and intention revision using the relativizing formulas that are created during the planning process. The system treats its speech acts just like its other actions. This general approach enables Eva to plan a variety of speech acts, including requests, informs, questions, confirmations, offers, acceptances, and emotive expressions. Because the dialogue engine is a planner, as the dialogue proceeds, the system can flexibly generate, execute, and potentially repair its plans using physical, digital, and speech actions. Importantly, Eva can explain its utterances because it has created a plan that caused it to utter them.

Zhuang Li, Levon Haroutunian, Raj Tumuluri et al.

Post-editing has proven effective in improving the quality of text generated by large language models (LLMs) such as GPT-3.5 or GPT-4, particularly when direct updating of their parameters to enhance text quality is infeasible or expensive. However, relying solely on smaller language models for post-editing can limit the LLMs' ability to generalize across domains. Moreover, the editing strategies in these methods are not optimally designed for text-generation tasks. To address these limitations, we propose a neural programmer-interpreter approach that preserves the domain generalization ability of LLMs when editing their output. The editing actions in this framework are specifically devised for text generation. Extensive experiments demonstrate that the programmer-interpreter significantly enhances GPT-3.5's performance in logical form-to-text conversion and low-resource machine translation, surpassing other state-of-the-art (SOTA) LLM post-editing methods in cross-domain settings.

Zhuang Li, Lizhen Qu, Philip R. Cohen et al.

Multilingual semantic parsing aims to leverage the knowledge from the high-resource languages to improve low-resource semantic parsing, yet commonly suffers from the data imbalance problem. Prior works propose to utilize the translations by either humans or machines to alleviate such issues. However, human translations are expensive, while machine translations are cheap but prone to error and bias. In this work, we propose an active learning approach that exploits the strengths of both human and machine translations by iteratively adding small batches of human translations into the machine-translated training set. Besides, we propose novel aggregated acquisition criteria that help our active learning method select utterances to be manually translated. Our experiments demonstrate that an ideal utterance selection can significantly reduce the error and bias in the translated data, resulting in higher parser accuracies than the parsers merely trained on the machine-translated data.

Philip R Cohen

This short position paper is intended to provide a critique of current approaches to dialogue, as well as a roadmap for collaborative dialogue research. It is unapologetically opinionated, but informed by 40 years of dialogue re-search. No attempt is made to be comprehensive. The paper will discuss current research into building so-called "chatbots", slot-filling dialogue systems, and plan-based dialogue systems. For further discussion of some of these issues, please see (Allen et al., in press).