72 lines
3.1 KiB
Markdown
72 lines
3.1 KiB
Markdown
|
---
|
|||
|
|
license: apache-2.0
|
|||
|
|
library_name: transformers
|
|||
|
|
pipeline_tag: text-generation
|
|||
|
|
language:
|
|||
|
|
- en
|
|||
|
|
tags:
|
|||
|
|
- math
|
|||
|
|
- reasoning
|
|||
|
|
- reasoning-compression
|
|||
|
|
- self-distillation
|
|||
|
|
- crisp
|
|||
|
|
base_model:
|
|||
|
|
- deepseek-ai/DeepSeek-R1-Distill-Llama-8B
|
|||
|
|
datasets:
|
|||
|
|
- pb09204048/CRISP
|
|||
|
|
---
|
|||
|
|
|
|||
|
|
# CRISP-DeepSeek-R1-Distill-Llama-8B-v2
|
|||
|
|
|
|||
|
|
**DeepSeek-R1-Distill-Llama-8B** trained with **CRISP** (Compressed Reasoning via Iterative Self-Policy Distillation)
|
|||
|
|
using the **v2** conciseness teacher. Step-99 checkpoint.
|
|||
|
|
|
|||
|
|
**Paper:** https://arxiv.org/abs/2603.05433
|
|||
|
|
|
|||
|
|
CRISP teaches a reasoning model to think concisely by distilling its own concise behavior back into
|
|||
|
|
itself: the teacher is the *same* model conditioned on a conciseness instruction, the student has no
|
|||
|
|
instruction, and training minimizes per-token reverse KL from student to teacher on the student's own
|
|||
|
|
rollouts (teacher refreshed every `M=50` steps). No ground-truth answers, token budgets, or
|
|||
|
|
difficulty estimators enter the loss.
|
|||
|
|
|
|||
|
|
This checkpoint uses the **v2** teacher prompt:
|
|||
|
|
**v2 (difficulty-aware, default):** adds a caveat to *not over-compress* hard/multi-step problems (keep case analysis, edge cases, a final check).
|
|||
|
|
|
|||
|
|
Other CRISP checkpoints: [Qwen3-8B](https://huggingface.co/pb09204048/CRISP-Qwen3-8B-v1) ([v2](https://huggingface.co/pb09204048/CRISP-Qwen3-8B-v2)),
|
|||
|
|
[Qwen3-14B](https://huggingface.co/pb09204048/CRISP-Qwen3-14B-v1) ([v2](https://huggingface.co/pb09204048/CRISP-Qwen3-14B-v2)),
|
|||
|
|
[DeepSeek-R1-Distill-Llama-8B](https://huggingface.co/pb09204048/CRISP-DeepSeek-R1-Distill-Llama-8B-v1) ([v2](https://huggingface.co/pb09204048/CRISP-DeepSeek-R1-Distill-Llama-8B-v2)).
|
|||
|
|
Training data: [pb09204048/CRISP](https://huggingface.co/datasets/pb09204048/CRISP).
|
|||
|
|
|
|||
|
|
## Usage
|
|||
|
|
|
|||
|
|
```python
|
|||
|
|
from transformers import AutoModelForCausalLM, AutoTokenizer
|
|||
|
|
tok = AutoTokenizer.from_pretrained("pb09204048/CRISP-DeepSeek-R1-Distill-Llama-8B-v2")
|
|||
|
|
model = AutoModelForCausalLM.from_pretrained("pb09204048/CRISP-DeepSeek-R1-Distill-Llama-8B-v2", device_map="auto")
|
|||
|
|
```
|
|||
|
|
|
|||
|
|
## Benchmark results (DeepSeek-R1-Distill-Llama-8B)
|
|||
|
|
|
|||
|
|
Accuracy (mean@8, %) and token reduction (Red., % vs. base) at a 30K-token budget. Math is scored with
|
|||
|
|
a dual-path grader (`Answer:` or `\boxed{}`); GPQA-Diamond and MMLU use exact letter-match. This
|
|||
|
|
model is the **CRISP (v2)** row.
|
|||
|
|
|
|||
|
|
| Setting | MATH-500 | AIME 2024 | AIME 2025 | GPQA-D | MMLU |
|
|||
|
|
|---------|----------|-----------|-----------|--------|------|
|
|||
|
|
| Base | 71.3 / — | 33.3 / — | 25.0 / — | 47.0 / — | 71.5 / — |
|
|||
|
|
| Concise prompt (v2) | 79.7 / 20.5% | 42.1 / 2.5% | 28.8 / 3.8% | 46.0 / 9.4% | 73.9 / 9.2% |
|
|||
|
|
| Concise prompt (v1) | 80.8 / 25.1% | 45.0 / 10.2% | 29.2 / 9.8% | 46.5 / 10.2% | 74.1 / 9.2% |
|
|||
|
|
| **CRISP (v2)** | 79.8 / 23.2% | 42.1 / −2.5% | 26.2 / 0.1% | 46.7 / 7.0% | 71.4 / 11.4% |
|
|||
|
|
| **CRISP (v1)** | 82.1 / 31.6% | 39.2 / 6.3% | 27.1 / 7.1% | 48.3 / 10.2% | 71.7 / 17.6% |
|
|||
|
|
|
|||
|
|
## Citation
|
|||
|
|
|
|||
|
|
```bibtex
|
|||
|
|
@article{sang2026crisp,
|
|||
|
|
title={Crisp: Compressed reasoning via iterative self-policy distillation},
|
|||
|
|
author={Sang, Hejian and Xu, Yuanda and Zhou, Zhengze and He, Ran and Wang, Zhipeng and Sun, Jiachen},
|
|||
|
|
journal={arXiv preprint arXiv:2603.05433},
|
|||
|
|
year={2026}
|
|||
|
|
}
|
|||
|
|
```
|