No official torrent yet — requesting it tells the flock what to host next. You cannot seed a catalog-only page.
Cataloged from vllm-sr/mmbert32k-feedback-detector-merged
All models by vllm-sr →A 4-class user feedback classifier based on mmbert-32k-yarn. This is the merged version with LoRA weights integrated - no PEFT library required.
This model classifies user messages into 4 feedback categories to help conversational AI systems understand user satisfaction:
Validation Results (2,985 samples):
Per-Class Performance:
Output:
This model was fine-tuned using LoRA (Low-Rank Adaptation) with the following configuration:
Supports 1800+ languages via Glot500 tokenizer. Best performance on:
Apache 2.0
from transformers import AutoModelForSequenceClassification, AutoTokenizerimport torch
# Load model and tokenizermodel = AutoModelForSequenceClassification.from_pretrained( "vllm-sr/mmbert32k-feedback-detector-merged")tokenizer = AutoTokenizer.from_pretrained( "vllm-sr/mmbert32k-feedback-detector-merged")model.eval()
# Label mappinglabels = ["SAT", "NEED_CLARIFICATION", "WRONG_ANSWER", "WANT_DIFFERENT"]
# Example inferencetexts = [ "Thank you, that's exactly what I needed!", "I don't understand, can you explain more?", "That's incorrect, the answer should be different.", "Can you give me another approach?",]
for text in texts: inputs = tokenizer(text, return_tensors="pt", truncation=True, max_length=512) with torch.no_grad(): outputs = model(**inputs) probs = torch.softmax(outputs.logits, dim=-1) pred = outputs.logits.argmax(-1).item() conf = probs[0][pred].item() print(f"{labels[pred]:20} ({conf:.1%}) | {text}")SAT (81.2%) | Thank you, that's exactly what I needed!NEED_CLARIFICATION (100.0%) | I don't understand, can you explain more?WRONG_ANSWER (100.0%) | That's incorrect, the answer should be different.WANT_DIFFERENT (100.0%) | Can you give me another approach?# Efficient batch processingtexts = ["Your text 1", "Your text 2", "Your text 3"]inputs = tokenizer(texts, return_tensors="pt", padding=True, truncation=True, max_length=512)
with torch.no_grad(): outputs = model(**inputs)
predictions = outputs.logits.argmax(-1).tolist()feedback_types = [labels[p] for p in predictions]@misc{mmbert32k-feedback-detector, title={mmBERT-32K Feedback Detector}, author={LLM Semantic Router Team}, year={2026}, publisher={Hugging Face}, url={https://huggingface.co/vllm-sr/mmbert32k-feedback-detector-merged}}