PimoLee5/qwen2.5-7b-cuad-qlora
text-generation model by PimoLee5
Older snapshot: check the original source for current information.
From the publisher
Model card & documentation
Source previewRead the publisher’s intended use, setup instructions, evaluations and limitations. The original model card is the source of truth.
A LoRA adapter for Qwen2.5-7B-Instruct that identifies contract clause types and answers "does this excerpt contain clause X, and what does it say?" with verbatim quotes from the excerpt. Trained with QLoRA on CUAD and evaluated before vs after on held-out contracts for accuracy, hallucination and calibration. Code, data pipeline and evaluation: https://github.com/raghavpli515/Legal-contract-LLM-fine-tuning-with-evaluation Demo video: https://youtu.be/hcN9TmknxCs 400 held-out items from 50 CUAD test contracts never seen in training. All runs use the same 4-bit quantization, greedy decoding and scorer. Brackets are 95% bootstrap CIs. "3-shot" gives the base model three worked examples of the same clause type in the prompt (no training). It helps Q&A partway and barely moves classification. Trade-off. Fine-tuning fixed recall and calibration: the base model says "not present" for most clauses that are there, usually with near-certainty. The cost is more wrong-clause-type answers on contracts without the clause. Those answers still quote real contract text, so…
Read the full model card ↗ · Preview checked 2026-11-12T04:23:08.877Z
Inside the original model card — Document outline
- Qwen2.5-7B-Instruct · CUAD contract-clause QLoRA adapter
- Results
- Intended use
- How to use
- Training
- Limitations
- Citation and licence
Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.
Documentation belongs to its respective authors. Reported project/model license: apache-2.0. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.
Model overview
PimoLee5/qwen2.5-7b-cuad-qlora is a text-generation model repository published by PimoLee5 on Hugging Face. The source reports the peft library.
This page summarizes Hub metadata. For intended use, training data, evaluation results and limitations, consult the original model card.
Model facts
- Task
- text-generation
- Library
- peft
- Recent downloads (30 days)
- 0
- Cumulative likes
- 0
- Hugging Face trending score
- 0
- Reported safetensors parameters
- Not reported
- Architecture
- Not reported
- License
- apache-2.0
- Access
- Not gated by Hugging Face
- Created
- 2026-10-03T06:37:24.000Z
- Last modified
- 2026-10-03T06:37:51.000Z
Compatibility and lineage
Reported languages: en
Reported base models: Qwen/Qwen2.5-7B-Instruct
Parameter count is not a RAM/VRAM requirement. Check precision, quantization, context length and runtime compatibility in the model card; no hardware or API-cost claim is inferred here.
Use and evaluate
Check license terms and access requirements first. Review the model card’s documented loading instructions, evaluations and safety limitations. Benchmark scores are not imported by this directory, and popularity is not an accuracy ranking.
No model weights are downloaded or executed by AltAPIs. Never enable remote model code without reviewing it.
Source and freshness
Source: Hugging Face Hub. Metadata observed 2026-10-03T06:39:34.535Z. Daily imports are snapshots, not real-time monitoring.
Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.