WildSEEK: Evaluating Language Models for Information-Seeking
Paper • 2608.30683 • Published
# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForSequenceClassification
tokenizer = AutoTokenizer.from_pretrained("tceron/info-seek-classifier")
model = AutoModelForSequenceClassification.from_pretrained("tceron/info-seek-classifier", device_map="auto")Quick Links
Model Card for Model ID
This classifier has been trained with ModernBERT large. It classifies user interactions with LLMs into 5 categories:
0: "information seeking",
1: "content creation",
2: "coding",
3: "not english",
4: "no request"
Model Details
All details in paper:
@inproceedings{ceron2026wildseek,
title = {WildSEEK: Evaluating Language Models for Information-Seeking},
author = {Ceron, Tanise and Baumann, Joachim and Bassignana, Elisa and Cabuk, Berat and Hovy, Dirk and Nozza, Debora},
booktitle = {Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing},
year = {2026}
}
Other classifiers from this study are:
tceron/high-risk-classifier— Other, Economic and Financial, Health, Politics, Judicial and Legal, Moral Values and Religion, Securitytceron/open-endedness-classifier— Analytical, Factoid
- Downloads last month
- 36
Model tree for tceron/info-seek-classifier
Base model
answerdotai/ModernBERT-large
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-classification", model="tceron/info-seek-classifier")