Position Debiasing Fine-Tuning for Causal Perception in Long-Term Dialogue
- URL: http://arxiv.org/abs/2406.02002v1
- Date: Tue, 4 Jun 2024 06:33:13 GMT
- Title: Position Debiasing Fine-Tuning for Causal Perception in Long-Term Dialogue
- Authors: Shixuan Fan, Wei Wei, Wendi Li, Xian-Ling Mao, Wenfeng Xie, Dangyang Chen,
- Abstract summary: The core of the dialogue system is to generate relevant, informative, and human-like responses based on extensive dialogue history.
There is a natural deficiency for such models, that is, inherent position bias.
We propose a novel method, named Causal Perception long-term Dialogue framework (CPD), which employs perturbation-based causal variable discovery method.
- Score: 26.171522810457486
- License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
- Abstract: The core of the dialogue system is to generate relevant, informative, and human-like responses based on extensive dialogue history. Recently, dialogue generation domain has seen mainstream adoption of large language models (LLMs), due to its powerful capability in generating utterances. However, there is a natural deficiency for such models, that is, inherent position bias, which may lead them to pay more attention to the nearby utterances instead of causally relevant ones, resulting in generating irrelevant and generic responses in long-term dialogue. To alleviate such problem, in this paper, we propose a novel method, named Causal Perception long-term Dialogue framework (CPD), which employs perturbation-based causal variable discovery method to extract casually relevant utterances from the dialogue history and enhances model causal perception during fine-tuning. Specifically, a local-position awareness method is proposed in CPD for inter-sentence position correlation elimination, which helps models extract causally relevant utterances based on perturbations. Then, a casual-perception fine-tuning strategy is also proposed, to enhance the capability of discovering the causal invariant factors, by differently perturbing causally relevant and non-casually relevant ones for response generation. Experimental results on two datasets prove that our proposed method can effectively alleviate the position bias for multiple LLMs and achieve significant progress compared with existing baselines.
Related papers
- Continual Dialogue State Tracking via Reason-of-Select Distillation [20.844176879145927]
We introduce the Reason-of-Select (RoS) distillation method by enhancing smaller models with a novel'meta-reasoning' capability.
The domain bootstrapping process enhances the model's ability to dissect intricate dialogues from multiple possible values.
Two novel improvements, "multi-value resolution" strategy and Semantic Contrastive Reasoning Selection method, significantly enhance RoS.
arXiv Detail & Related papers (2024-08-19T09:48:50Z) - PICK: Polished & Informed Candidate Scoring for Knowledge-Grounded
Dialogue Systems [59.1250765143521]
Current knowledge-grounded dialogue systems often fail to align the generated responses with human-preferred qualities.
We propose Polished & Informed Candidate Scoring (PICK), a generation re-scoring framework.
We demonstrate the effectiveness of PICK in generating responses that are more faithful while keeping them relevant to the dialogue history.
arXiv Detail & Related papers (2023-09-19T08:27:09Z) - EM Pre-training for Multi-party Dialogue Response Generation [86.25289241604199]
In multi-party dialogues, the addressee of a response utterance should be specified before it is generated.
We propose an Expectation-Maximization (EM) approach that iteratively performs the expectation steps to generate addressee labels.
arXiv Detail & Related papers (2023-05-21T09:22:41Z) - CausalDialogue: Modeling Utterance-level Causality in Conversations [83.03604651485327]
We have compiled and expanded upon a new dataset called CausalDialogue through crowd-sourcing.
This dataset includes multiple cause-effect pairs within a directed acyclic graph (DAG) structure.
We propose a causality-enhanced method called Exponential Average Treatment Effect (ExMATE) to enhance the impact of causality at the utterance level in training neural conversation models.
arXiv Detail & Related papers (2022-12-20T18:31:50Z) - DEAM: Dialogue Coherence Evaluation using AMR-based Semantic
Manipulations [46.942369532632604]
We propose a Dialogue Evaluation metric that relies on AMR-based semantic manipulations for incoherent data generation.
Our experiments show that DEAM achieves higher correlations with human judgments compared to baseline methods.
arXiv Detail & Related papers (2022-03-18T03:11:35Z) - Dialogue Inspectional Summarization with Factual Inconsistency Awareness [34.97845384948336]
We investigate the factual inconsistency problem for Dialogue Inspectional Summarization (DIS) under non-pretraining and pretraining settings.
An innovative end-to-end dialogue summary generation framework is proposed with two auxiliary tasks.
Comprehensive experiments demonstrate that the proposed model can generate a more readable summary with accurate coverage of factual aspects.
arXiv Detail & Related papers (2021-11-05T06:26:22Z) - Learning from Perturbations: Diverse and Informative Dialogue Generation
with Inverse Adversarial Training [10.17868476063421]
We propose Inverse Adversarial Training (IAT) algorithm for training neural dialogue systems.
IAT encourages the model to be sensitive to the perturbation in the dialogue history and therefore learning from perturbations.
We show that our approach can better model dialogue history and generate more diverse and consistent responses.
arXiv Detail & Related papers (2021-05-31T17:28:37Z) - I like fish, especially dolphins: Addressing Contradictions in Dialogue
Modeling [104.09033240889106]
We introduce the DialoguE COntradiction DEtection task (DECODE) and a new conversational dataset containing both human-human and human-bot contradictory dialogues.
We then compare a structured utterance-based approach of using pre-trained Transformer models for contradiction detection with the typical unstructured approach.
arXiv Detail & Related papers (2020-12-24T18:47:49Z) - Ranking Enhanced Dialogue Generation [77.8321855074999]
How to effectively utilize the dialogue history is a crucial problem in multi-turn dialogue generation.
Previous works usually employ various neural network architectures to model the history.
This paper proposes a Ranking Enhanced Dialogue generation framework.
arXiv Detail & Related papers (2020-08-13T01:49:56Z) - Dialogue-Based Relation Extraction [53.2896545819799]
We present the first human-annotated dialogue-based relation extraction (RE) dataset DialogRE.
We argue that speaker-related information plays a critical role in the proposed task, based on an analysis of similarities and differences between dialogue-based and traditional RE tasks.
Experimental results demonstrate that a speaker-aware extension on the best-performing model leads to gains in both the standard and conversational evaluation settings.
arXiv Detail & Related papers (2020-04-17T03:51:57Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.