# Joint Reasoning on Hybrid-knowledge sources for Task-Oriented Dialog

arXiv: https://arxiv.org/abs/2210.07295 · HTML: https://mlsanity.com/paper/2210.07295/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning.
- Builds on 21 of its 27 references that are in the map.
- First posted on arXiv 13 Oct 2022.

## Placed in

- LLM reasoning / [Reasoning benchmarks and evaluation](https://mlsanity.com/lr/reasoning-benchmarks-and-evaluation/)

## Builds on (most cited first, 21 of 21)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehension](https://mlsanity.com/paper/1910.13461/) (2019)
- [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing](https://mlsanity.com/paper/2107.13586/) (2021)
- [GPT Understands, Too](https://mlsanity.com/paper/2103.10385/) (2021)
- [ERNIE 3.0: Large-scale Knowledge Enhanced Pre-training for Language Understanding and Generation](https://mlsanity.com/paper/2107.02137/) (2021)
- [PTR: Prompt Tuning with Rules for Text Classification](https://mlsanity.com/paper/2105.11259/) (2021)
- [Key-Value Retrieval Networks for Task-Oriented Dialogue](https://mlsanity.com/paper/1705.05414/) (2017)
- [UBAR: Towards Fully End-to-End Task-Oriented Dialog Systems with GPT-2](https://mlsanity.com/paper/2012.03539/) (2020)
- [Learning End-to-End Goal-Oriented Dialog](https://mlsanity.com/paper/1605.07683/) (2016)
- [Frames: A Corpus for Adding Memory to Goal-Oriented Dialogue Systems](https://mlsanity.com/paper/1704.00057/) (2017)
- [Beyond Domain APIs: Task-oriented Conversational Modeling with Unstructured Knowledge Access](https://mlsanity.com/paper/2006.03533/) (2020)
- [The non-convex Burer-Monteiro approach works on smooth semidefinite programs](https://mlsanity.com/paper/1606.04970/) (2016)
- [Paraphrase Augmented Task-Oriented Dialog Generation](https://mlsanity.com/paper/2004.07462/) (2020)
- [Zero-shot Generalization in Dialog State Tracking through Generative Question Answering](https://mlsanity.com/paper/2101.08333/) (2021)
- [HybriDialogue: An Information-Seeking Dialogue Dataset Grounded on Tabular and Textual Data](https://mlsanity.com/paper/2204.13243/) (2022)
- [Simulated Chats for Building Dialog Systems: Learning to Generate Conversations from Instructions](https://mlsanity.com/paper/2010.10216/) (2020)
- [Constraint based Knowledge Base Distillation in End-to-End Task Oriented Dialogs](https://mlsanity.com/paper/2109.07396/) (2021)
- [Unsupervised Learning of KB Queries in Task-Oriented Dialogs](https://mlsanity.com/paper/2005.00123/) (2020)
- [HyKnow: End-to-End Task-Oriented Dialog Modeling with Hybrid Knowledge Management](https://mlsanity.com/paper/2105.06041/) (2021)
- [End-to-End Task-Oriented Dialog Modeling with Semi-Structured Knowledge Management](https://mlsanity.com/paper/2106.11796/) (2021)
- [A Knowledge-Grounded Dialog System Based on Pre-Trained Language Models](https://mlsanity.com/paper/2106.14444/) (2021)

## Abstract (arXiv)

Traditional systems designed for task oriented dialog utilize knowledge present only in structured knowledge sources to generate responses. However, relevant information required to generate responses may also reside in unstructured sources, such as documents. Recent state of the art models such as HyKnow and SeKnow aimed at overcoming these challenges make limiting assumptions about the knowledge sources. For instance, these systems assume that certain types of information, such as a phone number, is always present in a structured knowledge base (KB) while information about aspects such as entrance ticket prices, would always be available in documents. In this paper, we create a modified version of the MutliWOZ-based dataset prepared by SeKnow to demonstrate how current methods have significant degradation in performance when strict assumptions about the source of information are removed. Then, in line with recent work exploiting pre-trained language models, we fine-tune a BART based model using prompts for the tasks of querying knowledge sources, as well as, for response generation, without making assumptions about the information present in each knowledge source. Through a series of experiments, we demonstrate that our model is robust to perturbations to knowledge modality (source of information), and that it can fuse information from structured as well as unstructured knowledge to generate responses.