Young-Jun Lee

Requests for additional processed data (i.e., MedDialog-EN)

1

Hi, Can I share the additional processed data of MedDialog-EN? Thanks

Hi, I wonder that is it possible to control Bag-of-Words (i.e., introduced in PPLM paper)? If possible, then how to control it, and can you explain a description at the...

Execute interact.py

1

Hi, When I run `interact_adapter.py`, I used the below shell script. ```python #!/bin/bash python3 main.py -D sentiment --label_class 2 --length 30 --num_samples 10 --interact --verbose --sample_starter 10 --load_check_point_adapter runs/very_positive_0/pytorch_model.bin ```...

YFCC image path is not opened when reproducing Image-Chat model

1

Hi, I'm Young-Jun Lee. I'm currently reproducing the results of the image-chat model. I encountered a problem in that the YFCC image path is not opened (as shown in below...

Hi, It seems that the released code supports only a single input prompt during the inference stage. However, I want to perform batch inference for faster generation. Should I modify...

About the maximum length

1

Hi Young-Jin, What did you set the model's maximum length to? When I set it to 1024, it caused a CUDA out-of-memory issue. Best regards, Young-Jin

Hi, I want to run a proposed pipeline on my local server that functions similarly to https://wikichat.genie.stanford.edu/. When I input a query like 'Tell me about Red Velvet', I want...

How to print all pass@k scores when generating 16 samples?

10

Hi, I want to print all results of pass@k metrics when generating 16 samples. (e.g., k=1, 2, 4, 8, 16) ```python math_500_pass_k_at_16 = LightevalTaskConfig( name="math_500_pass_k_at_16", suite=["custom"], prompt_function=math_500_prompt_fn, hf_repo="HuggingFaceH4/MATH-500", hf_subset="default", hf_avail_splits=["test"],...

Hi, When running the GSM8K evaluation experiments using an 8-shot setting, I noticed that the few-shot examples were not successfully applied. Specifically, the current implementation triggers (in [here](https://github.com/QwenLM/Qwen2.5-Math/blob/a45202bd16f1ec06f433442dc1152d0074773465/evaluation/utils.py#L204C5-L214C59)). ```python if...