Instructions to use rishanthrajendhran/IdeaLens with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use rishanthrajendhran/IdeaLens with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="rishanthrajendhran/IdeaLens")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("rishanthrajendhran/IdeaLens")
model = AutoModelForCausalLM.from_pretrained("rishanthrajendhran/IdeaLens", device_map="auto")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use rishanthrajendhran/IdeaLens with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "rishanthrajendhran/IdeaLens"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rishanthrajendhran/IdeaLens",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/rishanthrajendhran/IdeaLens

SGLang

How to use rishanthrajendhran/IdeaLens with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "rishanthrajendhran/IdeaLens" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rishanthrajendhran/IdeaLens",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use 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 "rishanthrajendhran/IdeaLens" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "rishanthrajendhran/IdeaLens",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use rishanthrajendhran/IdeaLens with Docker Model Runner:
```
docker model run hf.co/rishanthrajendhran/IdeaLens
```

rishanthrajendhran commited on 5 days ago

Commit

9daf2d1

verified ·

1 Parent(s): 11f78dc

Rename the corpus IdeaLens-1M -> WildOutlines

Browse files

Files changed (2) hide show

README.md +6 -6
thresholds.json +1 -1

README.md CHANGED Viewed

@@ -10,7 +10,7 @@ tags:
 - ai-text-detection
 - idea-provenance
 datasets:
-- rishanthrajendhran/IdeaLens-1M
 extra_gated_prompt: "Access is granted individually. Please say who you are and what you intend to use the weights for."
 ---
@@ -21,7 +21,7 @@ role-labelled outline of the document (an ordered list of items, each giving one
 such as *Central Development* or *Open Question*) and returns P(human), the probability that the ideas are human.
 IdeaLens is `nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16` fine-tuned with LoRA (rank 64) on the outlines of 1M
-English web documents ([IdeaLens-1M](https://huggingface.co/datasets/rishanthrajendhran/IdeaLens-1M)). The training
 outlines were paraphrased to remove the documents' wording, so the model has to fit its labels through the ideas.
 ## Results
@@ -145,7 +145,7 @@ Scores agree with the training-time scores to about 0.001 in P(human) on average
 IdeaLens flags a document as having AI ideas when P(human) is below a cut. Each cut is set so
 that a given share of human documents is flagged (the false-positive rate, FPR), measured on the 80,000 human
-documents in IdeaLens-1M's `calibration` split (10,000 per format). The paper's operating point is the global
 cut at 1% FPR.
 | FPR | 0.1% | 0.5% | 1% | 2% | 5% | 10% | 20% |
@@ -154,7 +154,7 @@ cut at 1% FPR.
 Per-format cuts give each format its own operating point. They need the document's format, which the paper
 assigns with WebOrganizer's annotation prompt run on Gemini 3.7 Flash; the calibration documents use the formats
-recorded in IdeaLens-1M. Each is the
 format's own quantile, shrunk toward the global cut with weight n / (n + 2500); at 0.1% FPR 10,000 documents
 per format are too few, so there is no per-format cut. A document outside these eight formats has no
 per-format cut; do not fall back to the global cut for it.
@@ -198,7 +198,7 @@ input at a time:
 Inputs of 6,000 tokens do not fit on one 80 GB GPU and 12,000 do not fit on two; lowering the Mamba chunk size from
 128 to 64 did not change either limit.
-IdeaLens reads outlines, which are short. The outlines in IdeaLens-1M's calibration split average about 640
 tokens with the prompt, and the longest is under 3,800, so one 80 GB GPU (A100 80GB or H100 80GB) is enough. Outline
 extraction runs through an LLM API and needs no local GPU.
@@ -226,7 +226,7 @@ extraction runs through an LLM API and needs no local GPU.
 | [IdeaLens-ModernBERT-L-PerItem](https://huggingface.co/rishanthrajendhran/IdeaLens-ModernBERT-L-PerItem) | ModernBERT-large | single outline items, pooled |
 | [IdeaLens-LogisticClassifier-PerItem](https://huggingface.co/rishanthrajendhran/IdeaLens-LogisticClassifier-PerItem) | logistic regression over text-embedding-3-large | single outline items, pooled |
-Training data: [IdeaLens-1M](https://huggingface.co/datasets/rishanthrajendhran/IdeaLens-1M).
 ## License

 - ai-text-detection
 - idea-provenance
 datasets:
+- rishanthrajendhran/WildOutlines
 extra_gated_prompt: "Access is granted individually. Please say who you are and what you intend to use the weights for."
 ---
 such as *Central Development* or *Open Question*) and returns P(human), the probability that the ideas are human.
 IdeaLens is `nvidia/NVIDIA-Nemotron-3.5-Lightning-30B-A3B-BF16` fine-tuned with LoRA (rank 64) on the outlines of 1M
+English web documents ([WildOutlines](https://huggingface.co/datasets/rishanthrajendhran/WildOutlines)). The training
 outlines were paraphrased to remove the documents' wording, so the model has to fit its labels through the ideas.
 ## Results
 IdeaLens flags a document as having AI ideas when P(human) is below a cut. Each cut is set so
 that a given share of human documents is flagged (the false-positive rate, FPR), measured on the 80,000 human
+documents in WildOutlines's `calibration` split (10,000 per format). The paper's operating point is the global
 cut at 1% FPR.
 | FPR | 0.1% | 0.5% | 1% | 2% | 5% | 10% | 20% |
 Per-format cuts give each format its own operating point. They need the document's format, which the paper
 assigns with WebOrganizer's annotation prompt run on Gemini 3.7 Flash; the calibration documents use the formats
+recorded in WildOutlines. Each is the
 format's own quantile, shrunk toward the global cut with weight n / (n + 2500); at 0.1% FPR 10,000 documents
 per format are too few, so there is no per-format cut. A document outside these eight formats has no
 per-format cut; do not fall back to the global cut for it.
 Inputs of 6,000 tokens do not fit on one 80 GB GPU and 12,000 do not fit on two; lowering the Mamba chunk size from
 128 to 64 did not change either limit.
+IdeaLens reads outlines, which are short. The outlines in WildOutlines's calibration split average about 640
 tokens with the prompt, and the longest is under 3,800, so one 80 GB GPU (A100 80GB or H100 80GB) is enough. Outline
 extraction runs through an LLM API and needs no local GPU.
 | [IdeaLens-ModernBERT-L-PerItem](https://huggingface.co/rishanthrajendhran/IdeaLens-ModernBERT-L-PerItem) | ModernBERT-large | single outline items, pooled |
 | [IdeaLens-LogisticClassifier-PerItem](https://huggingface.co/rishanthrajendhran/IdeaLens-LogisticClassifier-PerItem) | logistic regression over text-embedding-3-large | single outline items, pooled |
+Training data: [WildOutlines](https://huggingface.co/datasets/rishanthrajendhran/WildOutlines).
 ## License

thresholds.json CHANGED Viewed

@@ -3,7 +3,7 @@
  "input": "outline",
  "flag_rule": "flag the document as AI when P(human) < cut",
  "calibration": {
-  "data": "IdeaLens-1M, calibration split (human documents only)",
   "scores_from": "the training checkpoint, scored on Tinker",
   "n_humans": 80000,
   "n_per_format": {

  "input": "outline",
  "flag_rule": "flag the document as AI when P(human) < cut",
  "calibration": {
+  "data": "WildOutlines, calibration split (human documents only)",
   "scores_from": "the training checkpoint, scored on Tinker",
   "n_humans": 80000,
   "n_per_format": {