sileod

tasksource. Datasets collection and preprocessings framework for NLP extreme multitask learning

tasknet. Easy modernBERT fine-tuning and multi-task learning

Discovery. Mining Discourse Markers for Unsupervised Sentence Representation Learning

reasoning-core. Procedural data generators for verifiable reasoning, synthetic pretraining, post-training, evaluation, and RL.

llm-theory-of-mind. Testing Theory of Mind (ToM) in language models with epistemic logic

pragmeval. Discourse Based Evaluation of Language Understanding

language-model-recommendation. Resources accompanying the "Zero-Shot Recommendation as Language Modeling" paper (ECIR2022)

DiscSense. Automated Semantic Analysis of Discourse Markers

xpflow. Python nested loops as classes for improved readability and modularity

metaeval. Collection of tasks for meta-learning and extreme multitask learning

gramforge. Efficient and multi-language generation from context free or sensitive grammars (CFG/CSG)

embcomp. Composition of embeddings

pandit. Pandas with some cool additional features

pygrove. Commons imports for data science using pyforest, with a focus on NLP

reasoning-core-openreward. Python

nlp-verbal-probabilities-reasoning. Probing handling of verbal probabilities in NLP models

litlm. Friendly and concise wrapper over litellm

nfsdict. NFS-safe persistent dictionary for Python