ronak commited on
Commit
f3a7cc4
1 Parent(s): 10b62a7

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +6 -6
README.md CHANGED
@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  # Model Card for RankZephyr 7B V1 - Full
18
 
19
- RankZephyr is a series of language models that are trained to act as helpful reranking assistants built on the Zephyr-7B-β model.
20
- RankZephyr Base is the model that follows single stage fine-tuning on the RankGPT-3.5 model, while RankZephyr Full is the model that is further fine-tuned on RankGPT-4 reorderings of OpenAI's Ada2 orderings for 5K queries.
21
 
22
 
23
  ## Model description
@@ -50,13 +50,13 @@ With the MS MARCO v1 collection:
50
  | RankGPT-3.5 | -| SPLADE++ ED | 0.7504 | 0.7120|
51
 
52
 
 
53
 
 
54
 
55
- ## Intended uses & limitationspacka
56
 
57
- The following is an excerpt from the [Zephyr-7B-β model card](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta/blob/main/README.md#intended-use--limitations):
58
-
59
- In our case, RankZephyr is fine-tuned to act as a listwise reranking agent. You provide it with a query and documents and get back a reordered list of document identifiers.
60
 
61
 
62
  ## Bias, Risks, and Limitations
 
16
 
17
  # Model Card for RankZephyr 7B V1 - Full
18
 
19
+ RankZephyr is a series of language models trained to act as helpful reranking assistants built on the Zephyr-7B-β model.
20
+ RankZephyr Base is the model that follows single-stage fine-tuning on the RankGPT-3.5 model, while RankZephyr Full is the model that is further fine-tuned on RankGPT-4 reorderings of OpenAI's Ada2 orderings for 5K queries.
21
 
22
 
23
  ## Model description
 
50
  | RankGPT-3.5 | -| SPLADE++ ED | 0.7504 | 0.7120|
51
 
52
 
53
+ More details can be found in the paper.
54
 
55
+ ## Intended uses & limitations
56
 
57
+ The model is to be used in conjunction with the [RankLLM repository](https://github.com/castorini/rank_llm). While `rank-llm` exists as a PyPI package, we are currently in the early stages of development and encourage users to directly check install from source.
58
 
59
+ The original Zephyr model is trained for chat. In our case, RankZephyr is fine-tuned to act as a listwise reranking agent. You provide it with a query and documents and get back a reordered list of document identifiers.
 
 
60
 
61
 
62
  ## Bias, Risks, and Limitations