Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userSPECTRA: Sparse Structured Text Rationalization
Spectra: ترشيد النص المنظم Sparse
175
0 0
0.0
(
0
)
Created by
Shamra Editor
Ask ChatGPT about the research
يهدف الترشيد الانتقائي إلى إنتاج قرارات جنبا إلى جنب مع المناولة (على سبيل المثال، النصوص النصية أو محاذاة الكلمات بين جملتين). عادة ما يتم تصميم المنظمات على غرار أقنعة ثنائية عشوائية، تتطلب من أدراج التدرج المستندة إلى أخذ العينات، والتي تعقد التدريب ويتطلب ضبط فرط الحرارة الدقيق. آليات الاهتمام المتنقل هي بديل محدد، لكنها تفتقر إلى طريقة لتنظيم استخراج الأساس المنطقي (على سبيل المثال، للسيطرة على Sparsity من تسليط الضوء على نص أو عدد المحاذاة). في هذه الورقة، نقدم إطارا موحدا لاستخراج التفسيرات المحددة من خلال الاستدلال المحدود على رسم بياني عامل، مما يشكل طبقة مختلفة. نهجنا تخفف إلى حد كبير التدريب والأساس المنطقي، مما يتفوق بشكل عام على العمل السابق بشأن ما يأتي إلى الأداء والمعقولية للمناولة المستخرجة. نحن نقدم كذلك دراسة مقارنة للأساليب الاستوكاستيكية والحتمية لاستخراج الأساس المنطقي لتصنيف مهام الاستدلال واللغة الطبيعية، وتقييم قوة التنبؤية بشكل مشترك، ونوعية التفسيرات، والتقلبات النموذجية.
Selective rationalization aims to produce decisions along with rationales (e.g., text highlights or word alignments between two sentences). Commonly, rationales are modeled as stochastic binary masks, requiring sampling-based gradient estimators, which complicates training and requires careful hyperparameter tuning. Sparse attention mechanisms are a deterministic alternative, but they lack a way to regularize the rationale extraction (e.g., to control the sparsity of a text highlight or the number of alignments). In this paper, we present a unified framework for deterministic extraction of structured explanations via constrained inference on a factor graph, forming a differentiable layer. Our approach greatly eases training and rationale regularization, generally outperforming previous work on what comes to performance and plausibility of the extracted rationales. We further provide a comparative study of stochastic and deterministic methods for rationale extraction for classification and natural language inference tasks, jointly assessing their predictive power, quality of the explanations, and model variability.
References used
https://aclanthology.org/
rate research
Read More
When learned without exploration, local models for structured prediction tasks are subject to exposure bias and cannot be trained without detailed guidance. Active Imitation Learning (AIL), also known in NLP as Dynamic Oracle Learning, is a general t
echnique for working around these issues by allowing the exploration of different outputs at training time. AIL requires oracle feedback: an oracle is any algorithm which can, given a partial candidate solution and gold annotation, find the correct (minimum loss) next output to produce. This paper describes a general finite state technique for deriving oracles. The technique describe is also efficient and will greatly expand the tasks for which AIL can be used.
Current sequence-to-sequence models are trained to minimize cross-entropy and use softmax to compute the locally normalized probabilities over target sequences. While this setup has led to strong results in a variety of tasks, one unsatisfying aspect
is its length bias: models give high scores to short, inadequate hypotheses and often make the empty string the argmax---the so-called cat got your tongue problem. Recently proposed entmax-based sparse sequence-to-sequence models present a possible solution, since they can shrink the search space by assigning zero probability to bad hypotheses, but their ability to handle word-level tasks with transformers has never been tested. In this work, we show that entmax-based models effectively solve the cat got your tongue problem, removing a major source of model error for neural machine translation. In addition, we generalize label smoothing, a critical regularization technique, to the broader family of Fenchel-Young losses, which includes both cross-entropy and the entmax losses. Our resulting label-smoothed entmax loss models set a new state of the art on multilingual grapheme-to-phoneme conversion and deliver improvements and better calibration properties on cross-lingual morphological inflection and machine translation for 7 language pairs.
Abstract Dual encoders perform retrieval by encoding documents and queries into dense low-dimensional vectors, scoring each document by its inner product with the query. We investigate the capacity of this architecture relative to sparse bag-of-words
models and attentional neural networks. Using both theoretical and empirical analysis, we establish connections between the encoding dimension, the margin between gold and lower-ranked documents, and the document length, suggesting limitations in the capacity of fixed-length encodings to support precise retrieval of long documents. Building on these insights, we propose a simple neural model that combines the efficiency of dual encoders with some of the expressiveness of more costly attentional architectures, and explore sparse-dense hybrids to capitalize on the precision of sparse retrieval. These models outperform strong alternatives in large-scale retrieval.
We present DART, an open domain structured DAta Record to Text generation dataset with over 82k instances (DARTs). Data-to-text annotations can be a costly process, especially when dealing with tables which are the major source of structured data and
contain nontrivial structures. To this end, we propose a procedure of extracting semantic triples from tables that encodes their structures by exploiting the semantic dependencies among table headers and the table title. Our dataset construction framework effectively merged heterogeneous sources from open domain semantic parsing and spoken dialogue systems by utilizing techniques including tree ontology annotation, question-answer pair to declarative sentence conversion, and predicate unification, all with minimum post-editing. We present systematic evaluation on DART as well as new state-of-the-art results on WebNLG 2017 to show that DART (1) poses new challenges to existing data-to-text datasets and (2) facilitates out-of-domain generalization. Our data and code can be found at https://github.com/Yale-LILY/dart.
The semantic matching capabilities of neural information retrieval can ameliorate synonymy and polysemy problems of symbolic approaches. However, neural models' dense representations are more suitable for re-ranking, due to their inefficiency. Sparse
representations, either in symbolic or latent form, are more efficient with an inverted index. Taking the merits of the sparse and dense representations, we propose an ultra-high dimensional (UHD) representation scheme equipped with directly controllable sparsity. UHD's large capacity and minimal noise and interference among the dimensions allow for binarized representations, which are highly efficient for storage and search. Also proposed is a bucketing method, where the embeddings from multiple layers of BERT are selected/merged to represent diverse linguistic aspects. We test our models with MS MARCO and TREC CAR, showing that our models outperforms other sparse models.
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
