中文
相关论文

相关论文: Semantic Sensitivities and Inconsistent Prediction…

200 篇论文

Success in natural language inference (NLI) should require a model to understand both lexical and compositional semantics. However, through adversarial evaluation, we find that several state-of-the-art models with diverse architectures are…

计算与语言 · 计算机科学 2018-11-20 Yixin Nie , Yicheng Wang , Mohit Bansal

Natural Language Inference is a challenging task that has received substantial attention, and state-of-the-art models now achieve impressive test set performance in the form of accuracy scores. Here, we go beyond this single evaluation…

计算与语言 · 计算机科学 2018-05-14 Vicente Ivan Sanchez Carmona , Jeff Mitchell , Sebastian Riedel

Do state-of-the-art models for language understanding already have, or can they easily learn, abilities such as boolean coordination, quantification, conditionals, comparatives, and monotonicity reasoning (i.e., reasoning about word…

计算与语言 · 计算机科学 2019-12-03 Kyle Richardson , Hai Hu , Lawrence S. Moss , Ashish Sabharwal

Rigorous evaluation of the causal effects of semantic features on language model predictions can be hard to achieve for natural language reasoning problems. However, this is such a desirable form of analysis from both an interpretability…

计算与语言 · 计算机科学 2024-04-04 Julia Rozanova , Marco Valentino , André Freitas

Rigorous evaluation of the causal effects of semantic features on language model predictions can be hard to achieve for natural language reasoning problems. However, this is such a desirable form of analysis from both an interpretability…

计算与语言 · 计算机科学 2024-04-04 Julia Rozanova , Marco Valentino , Andre Freitas

Standard evaluations of deep learning models for semantics using naturalistic corpora are limited in what they can tell us about the fidelity of the learned representations, because the corpora rarely come with good measures of semantic…

计算与语言 · 计算机科学 2018-11-01 Atticus Geiger , Ignacio Cases , Lauri Karttunen , Christopher Potts

Robustness evaluation for Natural Language to SQL (NL2SQL) systems is essential because real-world database environments are dynamic, noisy, and continuously evolving, whereas conventional benchmark evaluations typically assume static…

计算与语言 · 计算机科学 2026-03-19 Lifu Tu , Rongguang Wang , Tao Sheng , Sujjith Ravi , Dan Roth

The rapid advancement of Large Language Models (LLMs) has established standardized evaluation benchmarks as the primary instrument for model comparison. Yet, their reliability is increasingly questioned due to sensitivity to shallow…

计算与语言 · 计算机科学 2026-02-20 Bogdan Kostić , Conor Fallon , Julian Risch , Alexander Löser

In the interest of interpreting neural NLI models and their reasoning strategies, we carry out a systematic probing study which investigates whether these models capture the crucial semantic features central to natural logic: monotonicity…

计算与语言 · 计算机科学 2023-11-09 Julia Rozanova , Deborah Ferreira , Marco Valentino , Mokanrarangan Thayaparan , Andre Freitas

Current Natural Language Inference (NLI) models achieve impressive results, sometimes outperforming humans when evaluating on in-distribution test sets. However, as these models are known to learn from annotation artefacts and dataset…

计算与语言 · 计算机科学 2022-10-24 Joe Stacey , Pasquale Minervini , Haim Dubossarsky , Marek Rei

Natural Language Inference (NLI) has been a cornerstone task in evaluating language models' inferential reasoning capabilities. However, the standard three-way classification scheme used in NLI has well-known shortcomings in evaluating…

计算与语言 · 计算机科学 2023-06-19 Animesh Nighojkar , Antonio Laverghetta , John Licato

While many natural language inference (NLI) datasets target certain semantic phenomena, e.g., negation, tense & aspect, monotonicity, and presupposition, to the best of our knowledge, there is no NLI dataset that involves diverse types of…

计算与语言 · 计算机科学 2023-07-06 Lasha Abzianidze , Joost Zwarts , Yoad Winter

Small Language Models (SLMs) are increasingly being deployed in resource-constrained environments, yet their behavioral robustness to data contamination during instruction tuning remains poorly understood. We systematically investigate the…

计算与语言 · 计算机科学 2025-11-11 Nicy Scaria , Silvester John Joseph Kennedy , Deepak Subramani

Natural Language Inference (NLI) is the task of determining whether a sentence pair represents entailment, contradiction, or a neutral relationship. While NLI models perform well on many inference tasks, their ability to handle fine-grained…

计算与语言 · 计算机科学 2025-06-09 Tara Azin , Daniel Dumitrescu , Diana Inkpen , Raj Singh

Natural Language Inference (NLI) evaluation is crucial for assessing language understanding models; however, popular datasets suffer from systematic spurious correlations that artificially inflate actual model performance. To address this,…

计算与语言 · 计算机科学 2024-10-07 Adrian Cosma , Stefan Ruseti , Mihai Dascalu , Cornelia Caragea

Probing strategies have been shown to detect the presence of various linguistic features in large language models; in particular, semantic features intermediate to the "natural logic" fragment of the Natural Language Inference task (NLI).…

计算与语言 · 计算机科学 2023-04-21 Julia Rozanova , Marco Valentino , Lucas Cordeiro , Andre Freitas

We propose SETI (Systematicity Evaluation of Textual Inference), a novel and comprehensive benchmark designed for evaluating pre-trained language models (PLMs) for their systematicity capabilities in the domain of textual inference.…

计算与语言 · 计算机科学 2023-05-25 Xiyan Fu , Anette Frank

Natural Language Inference (NLI) models frequently rely on spurious correlations rather than semantic reasoning. Existing mitigation strategies often incur high annotation costs or trigger catastrophic forgetting during fine-tuning. We…

计算与语言 · 计算机科学 2025-12-23 Christopher Román Jaimes

Large language models exhibit systematic negation sensitivity, yet no operational framework exists to measure this vulnerability at deployment scale, especially in high-stakes decisions. We introduce Syntactic Framing Fragility (SFF), a…

计算与语言 · 计算机科学 2026-04-03 Katherine Elkins , Jon Chun

Large language models (LLMs) are increasingly applied in multilingual contexts, yet their capacity for consistent, logically grounded alignment across languages remains underexplored. We present a controlled evaluation framework for…

计算与语言 · 计算机科学 2025-08-21 Samir Abdaljalil , Erchin Serpedin , Khalid Qaraqe , Hasan Kurban
‹ 上一页 1 2 3 10 下一页 ›