Towards faithful model explanation in nlp: A survey
End-to-end neural Natural Language Processing (NLP) models are notoriously difficult to
understand. This has given rise to numerous efforts towards model explainability in recent …
understand. This has given rise to numerous efforts towards model explainability in recent …
Grammar-constrained decoding for structured NLP tasks without finetuning
Despite their impressive performance, large language models (LMs) still struggle with
reliably generating complex output structures when not finetuned to follow the required …
reliably generating complex output structures when not finetuned to follow the required …
Gluecons: A generic benchmark for learning under constraints
Recent research has shown that integrating domain knowledge into deep learning
architectures is effective; It helps reduce the amount of required data, improves the accuracy …
architectures is effective; It helps reduce the amount of required data, improves the accuracy …
Deal: Decoding-time alignment for large language models
Large Language Models (LLMs) are nowadays expected to generate content aligned with
human preferences. Current work focuses on alignment at model training time, through …
human preferences. Current work focuses on alignment at model training time, through …
Global constraints with prompting for zero-shot event argument classification
Determining the role of event arguments is a crucial subtask of event extraction. Most
previous supervised models leverage costly annotations, which is not practical for open …
previous supervised models leverage costly annotations, which is not practical for open …
Constraint reasoning embedded structured prediction
Many real-world structured prediction problems need machine learning to capture data
distribution and constraint reasoning to ensure structure validity. Nevertheless, constrained …
distribution and constraint reasoning to ensure structure validity. Nevertheless, constrained …
Long horizon forecasting with temporal point processes
In recent years, marked temporal point processes (MTPPs) have emerged as a powerful
modeling machinery to characterize asynchronous events in a wide variety of applications …
modeling machinery to characterize asynchronous events in a wide variety of applications …
Automata-based constraints for language model decoding
T Koo, F Liu, L He - arXiv preprint arXiv:2407.08103, 2024 - arxiv.org
LMs are often expected to generate strings in some formal language; for example, structured
data, API calls, or code snippets. Although LMs can be tuned to improve their adherence to …
data, API calls, or code snippets. Although LMs can be tuned to improve their adherence to …
[HTML][HTML] Unleashing the true potential of sequence-to-sequence models for sequence tagging and structure parsing
Abstract Sequence-to-Sequence (S2S) models have achieved remarkable success on
various text generation tasks. However, learning complex structures with S2S models …
various text generation tasks. However, learning complex structures with S2S models …
Long-Form Speech Translation through Segmentation with Finite-State Decoding Constraints on Large Language Models
AD McCarthy, H Zhang, S Kumar… - Findings of the …, 2023 - aclanthology.org
One challenge in speech translation is that plenty of spoken content is long-form, but short
units are necessary for obtaining high-quality translations. To address this mismatch, we …
units are necessary for obtaining high-quality translations. To address this mismatch, we …