HumorGen
Collection
Open-weight computational humor generation models including Core 7B suite, multilingual 14B/32B bases, and CLEF 2026 JOKER Task 4 models variants. • 15 items • Updated
How to use Jayi2424/HumorGen_JOKER_ES_32B with PEFT:
from peft import PeftModel
from transformers import AutoModelForCausalLM
base_model = AutoModelForCausalLM.from_pretrained("unsloth/qwen3-32b-bnb-4bit")
model = PeftModel.from_pretrained(base_model, "Jayi2424/HumorGen_JOKER_ES_32B")Part of the HumorGen Collection · SaLT Lab, Carnegie Mellon University
CLEF 2026 JOKER Task 4 constrained pun generation in Spanish. Two-stage cross-lingual LoRA curriculum on Qwen3-32B.
Paper(s): arXiv:2604.09629 · CLEF 2026 Working Notes
| Property | Value |
|---|---|
| Stage 1 | Multilingual pretraining — SemEval MWAHAHA |
| Stage 2 | CLEF-JOKER Task 4 — Spanish |
| Backbone | Qwen3-32B (QLoRA 4-bit) |
| LoRA r / alpha | 16 / 16 |
| Task | Dual-sense pun-brief generation |
This is a PEFT LoRA adapter. Load the base model and apply the adapter:
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen3-32B")
model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen3-32B", torch_dtype=torch.bfloat16, device_map="auto")
model = PeftModel.from_pretrained(model, "Jayi2424/HumorGen_JOKER_ES_32B")
pun_word = "bark"
sense_1 = "the sound a dog makes"
sense_2 = "the outer covering of a tree"
prompt = (
"<|im_start|>system\n"
"You are an expert at writing puns. Given a pun word and two meanings, write a "
"sentence that uses both senses naturally.\n<|im_end|>\n"
"<|im_start|>user\n"
f"Pun word: {pun_word}\nSense 1: {sense_1}\nSense 2: {sense_2}\n<|im_end|>\n"
"<|im_start|>assistant\n"
)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
outputs = model.generate(**inputs, max_new_tokens=80, temperature=0.8, top_p=0.95)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
@misc{ajayi2026humorgen,
title = {HumorGen: Cognitive Synergy for Humor Generation in Large Language
Models via Persona-Based Distillation},
author = {Ajayi, Edward and others},
year = {2026},
eprint = {2604.09629},
archivePrefix = {arXiv},
primaryClass = {cs.CL},
url = {https://arxiv.org/abs/2604.09629}
}
@inproceedings{ajayi2026joker,
title = {HumorGen at CLEF 2026 JOKER Task 4: Cross-Lingual Constrained
Pun Generation via the Cognitive Synergy Framework},
author = {Ajayi, Edward and others},
booktitle = {Working Notes of CLEF 2026},
year = {2026},
url = {https://edwardajayi.github.io/assets/papers/HumorGen-JOKER.pdf}
}
Base model
Qwen/Qwen3-32B