LaMP

LaMP copied to clipboard

Are we supposed to sort the dataset for Test data

1

As explained in the readme, we need to sort the data for test-data as well however while sorting the code tries to map the "output" parameter from the corresponding output...

mahima-umass

feebacks -> feedbacks

eltociear

Reproduction: about the specific Contriever model

1

Hi there! Thanks so much for your work on the LaMP benchmark. I've been working on reproducing the results but found myself a bit puzzled about the retriever. Facebook has...

JT-Ushio

Hi, thanks so much for your work of personalized LLM. I am currently trying to reproduce this work on my side. If anyone is also interested, might as well join...

Hannibal046

Why are the articles so shot? For the Personalized News Categorization dataset

1

Is text taken from the first sentence of the article ?

shuiba0

I tried to run the file RSPG/utils/create_data.py, and the parameter --retrievers_data_addr requires the presence of two files: scores.json and data.json. The scores.json file was obtained in the previous step, but...

Fang-git0

Below is an example prompt from User-based LaMP-1 using two arbitrarily retrieved profile: *For an author who has written the paper with the title, and "convex drawings of internally triconnected...

linfeng-du

why i can't access the data link

2

CUIH-77

In your evaluate_llm.py, you wrote: `opts = parser.parse_args() model = AutoModelForSeq2SeqLM.from_pretrained(opts.model_addr, cache_dir=opts.cache_dir) tokenizer = AutoTokenizer.from_pretrained(opts.model_addr, cache_dir=opts.cache_dir) collator = DataCollatorForSeq2Seq(tokenizer = tokenizer, model = model, max_length = opts.max_length)` But what if...

slz-ai

According to the README, this tool should use attribute "golds" but not the root attribute.

liaosunny123