Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userFine-grained Post-training for Improving Retrieval-based Dialogue Systems
بخير ما بعد التدريب على التدريب لتحسين أنظمة الحوار القائمة على الاسترجاع
249
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
تعرض أنظمة الحوار القائمة على استرجاع أداء متميز عند استخدام نماذج اللغة المدربة مسبقا، والتي تشمل تمثيلات تشفير ثنائية الاتجاه من المحولات (بيرت). خلال اختيار الاستجابة المتعدد التحويل، يركز بيرت على تدريب العلاقة بين السياق مع العديد من الكلام والاستجابة. ومع ذلك، فإن هذه الطريقة للتدريب غير كافية عند النظر في العلاقات بين كل كلام في السياق. هذا يؤدي إلى مشكلة عدم فهم تدفق السياق تماما المطلوب لتحديد استجابة. لمعالجة هذه المشكلة، نقترح طريقة جديدة بعد التدريب على ما بعد التدريب الجدد تعكس خصائص الحوار متعدد الدورات. على وجه التحديد، يتعلم النموذج تفاعلات مستوى الكلام من خلال التدريب على كل زوج استجابة سياق قصير في جلسة حوار. علاوة على ذلك، باستخدام هدف تدريب جديد، تصنيف صلة النطق، النموذج يفهم الأهمية الدلالية والتماسك بين كلام الحوار. تظهر النتائج التجريبية أن نموذجنا يحقق أحدث حديثة من بين الهوامش الهامة على ثلاثة مجموعات من مجموعات البيانات القياسية. هذا يشير إلى أن طريقة ما بعد التدريب الجيد غير فعالة للغاية لمهمة اختيار الاستجابة.
Retrieval-based dialogue systems display an outstanding performance when pre-trained language models are used, which includes bidirectional encoder representations from transformers (BERT). During the multi-turn response selection, BERT focuses on training the relationship between the context with multiple utterances and the response. However, this method of training is insufficient when considering the relations between each utterance in the context. This leads to a problem of not completely understanding the context flow that is required to select a response. To address this issue, we propose a new fine-grained post-training method that reflects the characteristics of the multi-turn dialogue. Specifically, the model learns the utterance level interactions by training every short context-response pair in a dialogue session. Furthermore, by using a new training objective, the utterance relevance classification, the model understands the semantic relevance and coherence between the dialogue utterances. Experimental results show that our model achieves new state-of-the-art with significant margins on three benchmark datasets. This suggests that the fine-grained post-training method is highly effective for the response selection task.
References used
https://aclanthology.org/
rate research
Read More
As the labeling cost for different modules in task-oriented dialog (ToD) systems is expensive, a major challenge is to train different modules with the least amount of labeled data. Recently, large-scale pre-trained language models, have shown promis
ing results for few-shot learning in ToD. In this paper, we devise a self-training approach to utilize the abundant unlabeled dialog data to further improve state-of-the-art pre-trained models in few-shot learning scenarios for ToD systems. Specifically, we propose a self-training approach that iteratively labels the most confident unlabeled data to train a stronger Student model. Moreover, a new text augmentation technique (GradAug) is proposed to better train the Student by replacing non-crucial tokens using a masked language model. We conduct extensive experiments and present analyses on four downstream tasks in ToD, including intent classification, dialog state tracking, dialog act prediction, and response selection. Empirical results demonstrate that the proposed self-training approach consistently improves state-of-the-art pre-trained models (BERT, ToD-BERT) when only a small number of labeled data are available.
In open-domain dialogue response generation, a dialogue context can be continued with diverse responses, and the dialogue models should capture such one-to-many relations. In this work, we first analyze the training objective of dialogue models from
the view of Kullback-Leibler divergence (KLD) and show that the gap between the real world probability distribution and the single-referenced data's probability distribution prevents the model from learning the one-to-many relations efficiently. Then we explore approaches to multi-referenced training in two aspects. Data-wise, we generate diverse pseudo references from a powerful pretrained model to build multi-referenced data that provides a better approximation of the real-world distribution. Model-wise, we propose to equip variational models with an expressive prior, named linear Gaussian model (LGM). Experimental results of automated evaluation and human evaluation show that the methods yield significant improvements over baselines.
This paper offers a comparative evaluation of four commercial ASR systems which are evaluated according to the post-editing effort required to reach publishable'' quality and according to the number of errors they produce. For the error annotation ta
sk, an original error typology for transcription errors is proposed. This study also seeks to examine whether there is a difference in the performance of these systems between native and non-native English speakers. The experimental results suggest that among the four systems, Trint obtains the best scores. It is also observed that most systems perform noticeably better with native speakers and that all systems are most prone to fluency errors.
Pre-trained Transformer language models (LM) have become go-to text representation encoders. Prior research fine-tunes deep LMs to encode text sequences such as sentences and passages into single dense vector representations for efficient text compar
ison and retrieval. However, dense encoders require a lot of data and sophisticated techniques to effectively train and suffer in low data situations. This paper finds a key reason is that standard LMs' internal attention structure is not ready-to-use for dense encoders, which needs to aggregate text information into the dense representation. We propose to pre-train towards dense encoder with a novel Transformer architecture, Condenser, where LM prediction CONditions on DENSE Representation. Our experiments show Condenser improves over standard LM by large margins on various text retrieval and similarity tasks.
Although exposure bias has been widely studied in some NLP tasks, it faces its unique challenges in dialogue response generation, the representative one-to-various generation scenario.In real human dialogue, there are many appropriate responses for t
he same context, not only with different expressions, but also with different topics. Therefore, due to the much bigger gap between various ground-truth responses and the generated synthetic response, exposure bias is more challenging in dialogue generation task.What's more, as MLE encourages the model to only learn the common words among different ground-truth responses, but ignores the interesting and specific parts, exposure bias may further lead to the common response generation problem, such as I don't know'' and HaHa?'' In this paper, we propose a novel adaptive switching mechanism, which learns to automatically transit between ground-truth learning and generated learning regarding the word-level matching score, such as the cosine similarity. Experimental results on both Chinese STC dataset and English Reddit dataset, show that our adaptive method achieves a significant improvement in terms of metric-based evaluation and human evaluation, as compared with the state-of-the-art exposure bias approaches. Further analysis on NMT task also shows that our model can achieve a significant improvement.
suggested questions
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
