arXiv:2610.19021v1 Announce Type: new Abstract: Recent work has questioned whether large language models (LLMs) can be useful for temporal event prediction. Despite vast pretrained world knowledge, L...
arXiv:2610.17113v1 Announce Type: cross Abstract: Information extraction models often perform well on the datasets they are trained on, but transferring to new domains and annotation guidelines remai...
The paper introduces a scalable cross‑domain event extraction system built on a unified generative sequence‑to‑sequence framework. It jointly handles event detection and argument extraction, allowing both pipeline and end‑to‑end configurations. By fine‑tuning pretrained language models on multiple event datasets from diverse domains, the system retains domain‑specific semantics while generalizing across large, evolving label spaces, and offers a web‑based application for researchers to upload documents, extract events, visualize triggers and arguments, and compare configurations.
arXiv:2608. 09209v1 Announce Type: cross Abstract: Neural language models trained on large crowdsourced corpora frequently exploit spurious surface patterns tied to target labels without true linguistic or causal relevance, boosting benchmark performance while failing on adversarial or out-of-distribution inputs.
arXiv:2510. 09783v2 Announce Type: replace-cross Abstract: Oversampling is one of the most widely used approaches for addressing imbalanced classification.
The paper introduces a weakly supervised framework for extracting dataset mentions from forced displacement and Fragile, Conflict, and Violence (FCV) documents. It uses a lightweight model trained on general research literature to generate candidate mentions, which are then refined by a large language model that validates or rejects them and corrects boundaries. The refined annotations are augmented with synthetic and contrastive examples to fine‑tune the model, achieving 74.1% precision and 70.5% recall on a benchmark of 1,706 passages, with higher precision (89.5%) on passages that contain dataset references.
SciTBERT is a family of BERT-derived language models trained on chronologically consistent corpora of scientific papers, patents, and educational web text, with training cutoffs spanning 2013 to 2025. The models are further post‑trained using paper and patent citations to create the SciTBERT‑CI family, which better captures temporal dynamics. Experiments show that these chronologically consistent encoders outperform earlier domain‑specific models, especially when training data is limited by early year restrictions, and they perform well on the new PatRepEval benchmark of patent‑related tasks at the science‑technology interface.
Neural language models trained on large crowdsourced corpora frequently exploit spurious surface patterns tied to target labels without true linguistic or causal relevance, boosting benchmark performance while failing on adversarial or out-of-distribution inputs. Existing approaches either require manual specification of the feature vocabulary or automate discovery only partially, leaving the gap between dataset-level correlation and model-level exploitation unaddressed.
arXiv:2609.07798v1 Announce Type: cross Abstract: Natural Language Processing (NLP) in the climate domain requires models to process heterogeneous text sources, including scientific literature, polic...
SAGE (Subpopulation-Aware Generative Enhancement) is a two-stage generative augmentation framework designed to mitigate spurious correlations in machine learning when group labels are unavailable. It uses cluster-derived sub-labels and class labels to fine‑tune a conditional generative model and text encoder, producing synthetic data that fills underrepresented regions and creates a balanced validation set for last‑layer reweighting. Experiments show SAGE improves worst‑group accuracy to 89.5%, 85.7%, and 79.1% on Waterbirds, CelebA, and MetaShift, outperforming existing group‑label‑free baselines by up to 7.7 percentage points.
arXiv:2606. 02991v1 Announce Type: cross Abstract: We introduce TypewriterLM, a 7.
SciNLP is a new benchmark dataset for full‑text entity and relation extraction in the NLP domain, comprising 60 manually annotated papers with 6,429 entities and 1,649 relations. It is the first dataset to provide full‑text annotations of entities and their relationships specifically for NLP literature. Experiments show that models trained on SciNLP outperform baselines on certain tasks, and the dataset enabled the automatic construction of a fine‑grained knowledge graph with an average node degree of 3.3.