arXiv subjects

Thibault Sellam

Publications and source records attributed to Thibault Sellam.

18 recordsLinked to original sources

arXiv2025

The FACTS Leaderboard: A Comprehensive Benchmark for Large Language Model Factuality

We introduce The FACTS Leaderboard, an online leaderboard suite and associated set of benchmarks that comprehensively evaluates the ability of language models to generate factually accurate text across diverse scenarios. The suite provides a holistic measure of factuality by aggregating the performance of models on four distinct sub-leaderboards: (1) FACTS Multimodal, which measures the factuality of responses to image-based questions; (2) FACTS Parametric, which assesses models' world knowledge by answering closed-book factoid questions from internal parameters; (3) FACTS Search, which evaluates factuality in information-seeking scenarios, where the model must use a search API; and (4) FACTS Grounding (v2), which evaluates whether long-form responses are grounded in provided documents, featuring significantly improved judge models. Each sub-leaderboard employs automated judge models to score model responses, and the final suite score is an average of the four components, designed to provide a robust and balanced assessment of a model's overall factuality. The FACTS Leaderboard Suite will be actively maintained, containing both public and private splits to allow for external participation while guarding its integrity. It can be found at https://www.kaggle.com/benchmarks/google/facts .

cs.CL↗

arXiv2023

SEAHORSE: A Multilingual, Multifaceted Dataset for Summarization Evaluation

Reliable automatic evaluation of summarization systems is challenging due to the multifaceted and subjective nature of the task. This is especially the case for languages other than English, where human evaluations are scarce. In this work, we introduce SEAHORSE, a dataset for multilingual, multifaceted summarization evaluation. SEAHORSE consists of 96K summaries with human ratings along 6 dimensions of text quality: comprehensibility, repetition, grammar, attribution, main ideas, and conciseness, covering 6 languages, 9 systems and 4 datasets. As a result of its size and scope, SEAHORSE can serve both as a benchmark to evaluate learnt metrics, as well as a large-scale resource for training such metrics. We show that metrics trained with SEAHORSE achieve strong performance on the out-of-domain meta-evaluation benchmarks TRUE (Honovich et al., 2022) and mFACE (Aharoni et al., 2022). We make the SEAHORSE dataset and metrics publicly available for future research on multilingual and multifaceted summarization evaluation.

cs.CL↗

arXiv2023

Reward Gaming in Conditional Text Generation

To align conditional text generation model outputs with desired behaviors, there has been an increasing focus on training the model using reinforcement learning (RL) with reward functions learned from human annotations. Under this framework, we identify three common cases where high rewards are incorrectly assigned to undesirable patterns: noise-induced spurious correlation, naturally occurring spurious correlation, and covariate shift. We show that even though learned metrics achieve high performance on the distribution of the data used to train the reward function, the undesirable patterns may be amplified during RL training of the text generation model. While there has been discussion about reward gaming in the RL or safety community, in this discussion piece, we would like to highlight reward gaming in the natural language generation (NLG) community using concrete conditional text generation examples and discuss potential fixes and areas for future work.

cs.CL↗

arXiv2023

SQuId: Measuring Speech Naturalness in Many Languages

Much of text-to-speech research relies on human evaluation, which incurs heavy costs and slows down the development process. The problem is particularly acute in heavily multilingual applications, where recruiting and polling judges can take weeks. We introduce SQuId (Speech Quality Identification), a multilingual naturalness prediction model trained on over a million ratings and tested in 65 locales-the largest effort of this type to date. The main insight is that training one model on many locales consistently outperforms mono-locale baselines. We present our task, the model, and show that it outperforms a competitive baseline based on w2v-BERT and VoiceMOS by 50.0%. We then demonstrate the effectiveness of cross-locale transfer during fine-tuning and highlight its effect on zero-shot locales, i.e., locales for which there is no fine-tuning data. Through a series of analyses, we highlight the role of non-linguistic effects such as sound artifacts in cross-locale transfer. Finally, we present the effect of our design decision, e.g., model size, pre-training diversity, and language rebalancing with several ablation experiments.

cs.CL↗

arXiv2022

Dialect-robust Evaluation of Generated Text

Evaluation metrics that are not robust to dialect variation make it impossible to tell how well systems perform for many groups of users, and can even penalize systems for producing text in lower-resource dialects. However, currently, there exists no way to quantify how metrics respond to change in the dialect of a generated utterance. We thus formalize dialect robustness and dialect awareness as goals for NLG evaluation metrics. We introduce a suite of methods and corresponding statistical tests one can use to assess metrics in light of the two goals. Applying the suite to current state-of-the-art metrics, we demonstrate that they are not dialect-robust and that semantic perturbations frequently lead to smaller decreases in a metric than the introduction of dialect features. As a first step to overcome this limitation, we propose a training schema, NANO, which introduces regional and language information to the pretraining process of a metric. We demonstrate that NANO provides a size-efficient way for models to improve the dialect robustness while simultaneously improving their performance on the standard metric benchmark.

cs.CL↗

arXiv2022

The MultiBERTs: BERT Reproductions for Robustness Analysis

Experiments with pre-trained models such as BERT are often based on a single checkpoint. While the conclusions drawn apply to the artifact tested in the experiment (i.e., the particular instance of the model), it is not always clear whether they hold for the more general procedure which includes the architecture, training data, initialization scheme, and loss function. Recent work has shown that repeating the pre-training process can lead to substantially different performance, suggesting that an alternate strategy is needed to make principled statements about procedures. To enable researchers to draw more robust conclusions, we introduce the MultiBERTs, a set of 25 BERT-Base checkpoints, trained with similar hyper-parameters as the original BERT model but differing in random weight initialization and shuffling of training data. We also define the Multi-Bootstrap, a non-parametric bootstrap method for statistical inference designed for settings where there are multiple pre-trained models and limited test data. To illustrate our approach, we present a case study of gender bias in coreference resolution, in which the Multi-Bootstrap lets us measure effects that may not be detected with a single checkpoint. We release our models and statistical library along with an additional set of 140 intermediate checkpoints captured during pre-training to facilitate research on learning dynamics.

cs.CL↗

arXiv2022

Repairing the Cracked Foundation: A Survey of Obstacles in Evaluation Practices for Generated Text

Evaluation practices in natural language generation (NLG) have many known flaws, but improved evaluation approaches are rarely widely adopted. This issue has become more urgent, since neural NLG models have improved to the point where they can often no longer be distinguished based on the surface-level features that older metrics rely on. This paper surveys the issues with human and automatic model evaluations and with commonly used datasets in NLG that have been pointed out over the past 20 years. We summarize, categorize, and discuss how researchers have been addressing these issues and what their findings mean for the current state of model evaluations. Building on those insights, we lay out a long-term vision for NLG evaluation and propose concrete steps for researchers to improve their evaluation processes. Finally, we analyze 66 NLG papers from recent NLP conferences in how well they already follow these suggestions and identify which areas require more drastic changes to the status quo.

cs.CL↗

arXiv2021

Learning Compact Metrics for MT

Recent developments in machine translation and multilingual text generation have led researchers to adopt trained metrics such as COMET or BLEURT, which treat evaluation as a regression problem and use representations from multilingual pre-trained models such as XLM-RoBERTa or mBERT. Yet studies on related tasks suggest that these models are most efficient when they are large, which is costly and impractical for evaluation. We investigate the trade-off between multilinguality and model capacity with RemBERT, a state-of-the-art multilingual language model, using data from the WMT Metrics Shared Task. We present a series of experiments which show that model size is indeed a bottleneck for cross-lingual transfer, then demonstrate how distillation can help addressing this bottleneck, by leveraging synthetic data generation and transferring knowledge from one teacher to multiple students trained on related languages. Our method yields up to 10.5% improvement over vanilla fine-tuning and reaches 92.6% of RemBERT's performance using only a third of its parameters.

cs.CL↗

arXiv2021

The GEM Benchmark: Natural Language Generation, its Evaluation and Metrics

We introduce GEM, a living benchmark for natural language Generation (NLG), its Evaluation, and Metrics. Measuring progress in NLG relies on a constantly evolving ecosystem of automated metrics, datasets, and human evaluation standards. Due to this moving target, new models often still evaluate on divergent anglo-centric corpora with well-established, but flawed, metrics. This disconnect makes it challenging to identify the limitations of current models and opportunities for progress. Addressing this limitation, GEM provides an environment in which models can easily be applied to a wide set of tasks and in which evaluation strategies can be tested. Regular updates to the benchmark will help NLG research become more multilingual and evolve the challenge alongside models. This paper serves as the description of the data for which we are organizing a shared task at our ACL 2021 Workshop and to which we invite the entire NLG community to participate.

cs.CL↗

arXiv2020

Sticking to the Facts: Confident Decoding for Faithful Data-to-Text Generation

We address the issue of hallucination in data-to-text generation, i.e., reducing the generation of text that is unsupported by the source. We conjecture that hallucination can be caused by an encoder-decoder model generating content phrases without attending to the source; so we propose a confidence score to ensure that the model attends to the source whenever necessary, as well as a variational Bayes training framework that can learn the score from data. Experiments on the WikiBio (Lebretet al., 2016) dataset show that our approach is more faithful to the source than existing state-of-the-art approaches, according to both PARENT score (Dhingra et al., 2019) and human evaluation. We also report strong results on the WebNLG (Gardent et al., 2017) dataset.

cs.CL↗

arXiv2020

Learning to Evaluate Translation Beyond English: BLEURT Submissions to the WMT Metrics 2020 Shared Task

The quality of machine translation systems has dramatically improved over the last decade, and as a result, evaluation has become an increasingly challenging problem. This paper describes our contribution to the WMT 2020 Metrics Shared Task, the main benchmark for automatic evaluation of translation. We make several submissions based on BLEURT, a previously published metric based on transfer learning. We extend the metric beyond English and evaluate it on 14 language pairs for which fine-tuning data is available, as well as 4 "zero-shot" language pairs, for which we have no labelled examples. Additionally, we focus on English to German and demonstrate how to combine BLEURT's predictions with those of YiSi and use alternative reference translations to enhance the performance. Empirical results show that the models achieve competitive results on the WMT Metrics 2019 Shared Task, indicating their promise for the 2020 edition.

cs.CL↗

arXiv2020

A Multilingual View of Unsupervised Machine Translation

We present a probabilistic framework for multilingual neural machine translation that encompasses supervised and unsupervised setups, focusing on unsupervised translation. In addition to studying the vanilla case where there is only monolingual data available, we propose a novel setup where one language in the (source, target) pair is not associated with any parallel data, but there may exist auxiliary parallel data that contains the other. This auxiliary data can naturally be utilized in our probabilistic framework via a novel cross-translation loss term. Empirically, we show that our approach results in higher BLEU scores over state-of-the-art unsupervised models on the WMT'14 English-French, WMT'16 English-German, and WMT'16 English-Romanian datasets in most directions. In particular, we obtain a +1.65 BLEU advantage over the best-performing unsupervised model in the Romanian-English direction.

cs.CL↗

arXiv2020

BLEURT: Learning Robust Metrics for Text Generation

Text generation has made significant advances in the last few years. Yet, evaluation metrics have lagged behind, as the most popular choices (e.g., BLEU and ROUGE) may correlate poorly with human judgments. We propose BLEURT, a learned evaluation metric based on BERT that can model human judgments with a few thousand possibly biased training examples. A key aspect of our approach is a novel pre-training scheme that uses millions of synthetic examples to help the model generalize. BLEURT provides state-of-the-art results on the last three years of the WMT Metrics shared task and the WebNLG Competition dataset. In contrast to a vanilla BERT-based approach, it yields superior results even when the training data is scarce and out-of-distribution.

cs.CL↗

arXiv2019

Mining Precision Interfaces From Query Logs

Interactive tools make data analysis more efficient and more accessible to end-users by hiding the underlying query complexity and exposing interactive widgets for the parts of the query that matter to the analysis. However, creating custom tailored (i.e., precise) interfaces is very costly, and automated approaches are desirable. We propose a syntactic approach that uses queries from an analysis to generate a tailored interface. We model interface widgets as functions I(q) -> q' that modify the current analysis query $q$, and interfaces as the set of queries that its widgets can express. Our system, Precision Interfaces, analyzes structural changes between input queries from an analysis, and generates an output interface with widgets to express those changes. Our experiments on the Sloan Digital Sky Survey query log suggest that Precision Interfaces can generate useful interfaces for simple unanticipated tasks, and our optimizations can generate interfaces from logs of up to 10,000 queries in <10s.

cs.DB↗

arXiv2019

DeepBase: Deep Inspection of Neural Networks

Although deep learning models perform remarkably well across a range of tasks such as language translation and object recognition, it remains unclear what high-level logic, if any, they follow. Understanding this logic may lead to more transparency, better model design, and faster experimentation. Recent machine learning research has leveraged statistical methods to identify hidden units that behave (e.g., activate) similarly to human understandable logic, but those analyses require considerable manual effort. Our insight is that many of those studies follow a common analysis pattern, which we term Deep Neural Inspection. There is opportunity to provide a declarative abstraction to easily express, execute, and optimize them. This paper describes DeepBase, a system to inspect neural network behaviors through a unified interface. We model logic with user-provided hypothesis functions that annotate the data with high-level labels (e.g., part-of-speech tags, image captions). DeepBase lets users quickly identify individual or groups of units that have strong statistical dependencies with desired hypotheses. We discuss how DeepBase can express existing analyses, propose a set of simple and effective optimizations to speed up a standard Python implementation by up to 72x, and reproduce recent studies from the NLP literature.

cs.DB↗

arXiv2017

Mining Precision Interfaces From Query Logs

Interactive tools make data analysis both more efficient and more accessible to a broad population. Simple interfaces such as Google Finance as well as complex visual exploration interfaces such as Tableau are effective because they are tailored to the desired user tasks. Yet, designing interactive interfaces requires technical expertise and domain knowledge. Experts are scarce and expensive, and therefore it is currently infeasible to provide tailored (or precise) interfaces for every user and every task. We envision a data-driven approach to generate tailored interactive interfaces. We observe that interactive interfaces are designed to express sets of programs; thus, samples of programs-increasingly collected by data systems-may help us build interactive interfaces. Based on this idea, Precision Interfaces is a language-agnostic system that examines an input query log, identifies how the queries structurally change, and generates interactive web interfaces to express these changes. The focus of this paper is on applying this idea towards logs of structured queries. Our experiments show that Precision Interfaces can support multiple query languages (SQL and SPARQL), derive Tableau's salient interaction components from OLAP queries, analyze <75k queries in <12 minutes, and generate interaction designs that improve upon existing interfaces and are comparable to human-crafted interfaces.

cs.DB↗

arXiv2017

Precision Interfaces

Building interactive tools to support data analysis is hard because it is not always clear what to build and how to build it. To address this problem, we present Precision Interfaces, a semi-automatic system to generate task-specific data analytics interfaces. Precision Interface can turn a log of executed programs into an interface, by identifying micro-variations between the programs and mapping them to interface components. This paper focuses on SQL query logs, but we can generalize the approach to other languages. Our system operates in two steps: it first build an interaction graph, which describes how the queries can be transformed into each other. Then, it finds a set of UI components that covers a maximal number of transformations. To restrict the domain of changes to be detected, our system uses a domain-specific language, PILang. We give a full description of Precision Interface's components, showcase an early prototype on real program logs and discuss future research opportunities.

cs.DB↗

arXiv2017

80 New Packages to Mine Database Query Logs

The query log of a DBMS is a powerful resource. It enables many practical applications, including query optimization and user experience enhancement. And yet, mining SQL queries is a difficult task. The fundamental problem is that queries are symbolic objects, not vectors of numbers. Therefore, many popular statistical concepts, such as means, regression, or decision trees do not apply. Most authors limit themselves to ad hoc algorithms or approaches based on neighborhoods, such as k Nearest Neighbors. Our project is to challenge this limitation. We introduce methods to manipulate SQL queries as if they were vectors, thereby unlocking the whole statistical toolbox. We present three families of methods: feature maps, kernel methods, and Bayesian models. The first technique directly encodes queries into vectors. The second one transforms the queries implicitly. The last one exploits probabilistic graphical models as an alternative to vector spaces. We present the benefits and drawbacks of each solution, highlight how they relate to each other, and make the case for future investigation.

cs.DB↗