DDRO-Reference-Policies (SFT)
Collection
Step-2 SFT reference policies (ฯ_ref) used to initialize DDRO (MS MARCO / NQ; PQ and Title+URL DocIDs); use these for fair comparisons/ablations. โข 4 items โข Updated
Text Generation
Transformers
Safetensors
English
t5
text2text-generation
generative-retrieval
information-retrieval
ddro
sft
reference-policy
text-generation-inference
How to use from
SGLangUse Docker images
docker run --gpus all \
--shm-size 32g \
-p 30000:30000 \
-v ~/.cache/huggingface:/root/.cache/huggingface \
--env "HF_TOKEN=<secret>" \
--ipc=host \
lmsysorg/sglang:latest \
python3 -m sglang.launch_server \
--model-path "kiyam/ddro-nq-pq-sft" \
--host 0.0.0.0 \
--port 30000# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
-H "Content-Type: application/json" \
--data '{
"model": "kiyam/ddro-nq-pq-sft",
"prompt": "Once upon a time,",
"max_tokens": 512,
"temperature": 0.5
}'Quick Links
DDRO Step-2 SFT Reference Policy โ NQ + PQ
This repository contains the Step 2 (SFT) checkpoint used as the reference policy (ฯ_ref) / initialization for the DDRO training stage from Lightweight and Direct Document Relevance Optimization for Generative Information Retrieval.
What this is
- Stage: Step 2 / supervised fine-tuning (SFT)
- Role in DDRO: reference policy (ฯ_ref)
Corresponding final DDRO model
- Post-DDRO (Step 3) checkpoint:
kiyam/ddro-nq-pq
How to load
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
repo = "kiyam/ddro-nq-pq-sft"
tok = AutoTokenizer.from_pretrained(repo)
model = AutoModelForSeq2SeqLM.from_pretrained(repo)
Intended use
Research and reproducibility. Please cite the DDRO paper if you use this checkpoint.
- Downloads last month
- 6
Install from pip and serve model
# Install SGLang from pip: pip install sglang# Start the SGLang server: python3 -m sglang.launch_server \ --model-path "kiyam/ddro-nq-pq-sft" \ --host 0.0.0.0 \ --port 30000# Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "kiyam/ddro-nq-pq-sft", "prompt": "Once upon a time,", "max_tokens": 512, "temperature": 0.5 }'