初始化项目，由ModelHub XC社区提供模型

Model: pymlex/Qwen2.5-0.5B-Human Source: Original Platform
2026-07-07 20:13:26 +08:00
commit f9d9d3a8a6
17 changed files with 368 additions and 0 deletions
--- a/.gitattributes
+++ b/.gitattributes
@@ -0,0 +1,38 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text
+assets/training_monitor_analysis.png filter=lfs diff=lfs merge=lfs -text
+assets/training_summary.png filter=lfs diff=lfs merge=lfs -text
--- a/README.md
+++ b/README.md
@@ -0,0 +1,107 @@
+---
+license: gpl-3.0
+language:
+- es
+base_model: Qwen/Qwen2.5-0.5B-Instruct
+tags:
+- dpo
+- ai-detection
+- paraphrase
+library_name: transformers
+datasets:
+- pymlex/ai-generated-texts
+metrics:
+- accuracy
+pipeline_tag: text-generation
+---
+
+# Qwen2.5-0.5B-Human
+
+DPO fine-tune of [Qwen/Qwen2.5-0.5B-Instruct](https://huggingface.co/Qwen/Qwen2.5-0.5B-Instruct) that paraphrases Spanish academic abstracts to reduce AI-detection scores from [danibor/oculus-v2.0-multilingual](https://huggingface.co/danibor/oculus-v2.0-multilingual).
+
+## Overview
+
+Preference pairs for optimisation come from [pymlex/ai-generated-texts](https://huggingface.co/datasets/pymlex/ai-generated-texts). The corpus is [Flaglab/academic-knowledge-abstracts-es](https://huggingface.co/datasets/Flaglab/academic-knowledge-abstracts-es). For each train abstract, two base-model paraphrases are ranked by Oculus logit. DPO with beta = 0.1 increases the relative log-probability of the lower-logit completion. Retained pairs: 6396 from 8891 train abstracts with absolute logit gap at least 1.
+
+## Evaluation setup
+
+Hardware: NVIDIA RTX 5090, Ubuntu Jupyter, CUDA 13.0+, bf16 training and inference. Post-training evaluation generates one paraphrase per validation and test abstract with the base and fine-tuned models, scores each output with Oculus, and treats label 1 as AI-generated at threshold 0.5 on detector probability.
+
+During DPO, mean validation AI probability on a 276-text subset moved from 0.6740 at step 0 to 0.2437 at the last monitor step (-0.4303).
+
+## Results
+
+| Model | Split | n | mean prob | mean logit | accuracy | MCC | ROC-AUC | F1 |
+| --- | --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: |
+| base | validation | 1107 | 0.6550 | 1.4619 | 0.6712 | 0.0000 | n/a | 0.8032 |
+| base | test | 1112 | 0.6532 | 1.5581 | 0.6655 | 0.0000 | n/a | 0.7991 |
+| fine-tuned | validation | 1107 | 0.2264 | -2.0100 | 0.1716 | 0.0000 | n/a | 0.2930 |
+| fine-tuned | test | 1112 | 0.2391 | -1.8733 | 0.1835 | 0.0000 | n/a | 0.3100 |
+
+Lower mean probability and MCC near zero indicate weaker detector response on model paraphrases under the AI-positive labelling convention.
+
+![Evaluation summary](https://huggingface.co/pymlex/Qwen2.5-0.5B-Human/resolve/main/assets/evaluation_summary.png)
+
+![Score distributions](https://huggingface.co/pymlex/Qwen2.5-0.5B-Human/resolve/main/assets/score_distributions.png)
+
+![Training monitor](https://huggingface.co/pymlex/Qwen2.5-0.5B-Human/resolve/main/assets/training_monitor_analysis.png)
+
+## Source code
+
+The full pipeline is published on [GitHub](https://github.com/pymlex/ai-text-detector-tricking).
+
+## Citation
+
+If you found this model useful, please cite it as:
+
+```bibtex
+@misc{zyukov2026qwenhuman,
+  title         = {{Qwen2.5-0.5B-Human: DPO fine-tune against Oculus detector}},
+  author        = {Zyukov, Alex},
+  year          = {2026},
+  url           = {https://huggingface.co/pymlex/Qwen2.5-0.5B-Human}
+}
+```
+
+```bibtex
+@misc{zyukov2026aitexttricking,
+  title         = {{DPO Fine-Tuning Against Multilingual AI Text Detectors}},
+  author        = {Zyukov, Alex},
+  year          = {2026},
+  url           = {https://github.com/pymlex/ai-text-detector-tricking},
+  publisher     = {GitHub},
+  organization  = {pymlex}
+}
+```
+
+```bibtex
+@misc{nicks2024detectors,
+  title         = {{Language Model Detectors Are Easily Optimized Against}},
+  author        = {Nicks, Cameron and Chua, Jeremy and Liu, Stephen and others},
+  year          = {2024},
+  eprint        = {2406.07490},
+  archivePrefix = {arXiv},
+  primaryClass  = {cs.CL},
+  url           = {https://arxiv.org/abs/2406.07490}
+}
+```
+
+```bibtex
+@misc{oculus2026,
+  title         = {{Oculus 2.0 Multilingual AI Text Detector}},
+  author        = {danibor},
+  year          = {2026},
+  url           = {https://huggingface.co/danibor/oculus-v2.0-multilingual}
+}
+```
+
+```bibtex
+@misc{flaglab2025abstracts,
+  title         = {{Academic Knowledge Abstracts Spanish}},
+  author        = {Flaglab},
+  year          = {2025},
+  url           = {https://huggingface.co/datasets/Flaglab/academic-knowledge-abstracts-es}
+}
+```
+
+The project is under GPL-3.0 license.
--- a/assets/base_evaluation_report.json
+++ b/assets/base_evaluation_report.json
@@ -0,0 +1,27 @@
+{
+  "validation": {
+    "split": "validation",
+    "n_samples": 1107,
+    "accuracy": 0.6711833785004516,
+    "precision": 1.0,
+    "recall": 0.6711833785004516,
+    "f1": 0.8032432432432433,
+    "mcc": 0.0,
+    "roc_auc": null,
+    "mean_logit": 1.4619352543295816,
+    "mean_probability": 0.6550314651811926
+  },
+  "test": {
+    "split": "test",
+    "n_samples": 1112,
+    "accuracy": 0.6654676258992805,
+    "precision": 1.0,
+    "recall": 0.6654676258992805,
+    "f1": 0.7991360691144709,
+    "mcc": 0.0,
+    "roc_auc": null,
+    "mean_logit": 1.5581493150792922,
+    "mean_probability": 0.653164010906495
+  },
+  "threshold": 0.5
+}
--- a/assets/detector_probability_hist_test.png
+++ b/assets/detector_probability_hist_test.png
--- a/assets/detector_probability_hist_validation.png
+++ b/assets/detector_probability_hist_validation.png
--- a/assets/evaluation_report.json
+++ b/assets/evaluation_report.json
@@ -0,0 +1,27 @@
+{
+  "validation": {
+    "split": "validation",
+    "n_samples": 1107,
+    "accuracy": 0.17163504968383017,
+    "precision": 1.0,
+    "recall": 0.17163504968383017,
+    "f1": 0.29298380878951424,
+    "mcc": 0.0,
+    "roc_auc": null,
+    "mean_logit": -2.0099520581162786,
+    "mean_probability": 0.2263815667345318
+  },
+  "test": {
+    "split": "test",
+    "n_samples": 1112,
+    "accuracy": 0.18345323741007194,
+    "precision": 1.0,
+    "recall": 0.18345323741007194,
+    "f1": 0.3100303951367781,
+    "mcc": 0.0,
+    "roc_auc": null,
+    "mean_logit": -1.8733430301172578,
+    "mean_probability": 0.23908082744766893
+  },
+  "threshold": 0.5
+}
--- a/assets/evaluation_summary.png
+++ b/assets/evaluation_summary.png
--- a/assets/score_distributions.png
+++ b/assets/score_distributions.png
--- a/assets/training_monitor_analysis.png
+++ b/assets/training_monitor_analysis.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:65481b02b5cdb712ab905c8de2b38e602f0c529cd12ad523c98571f0a22d7871
+size 107297
--- a/assets/training_summary.png
+++ b/assets/training_summary.png
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:17a7da8f8f252d87024cb2292c642c68964c8099229eea4e73fb9cb47de32041
+size 108928
--- a/chat_template.jinja
+++ b/chat_template.jinja
@@ -0,0 +1,54 @@
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- messages[0]['content'] }}
+    {%- else %}
+        {{- 'You are Qwen, created by Alibaba Cloud. You are a helpful assistant.' }}
+    {%- endif %}
+    {{- "\n\n# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0]['role'] == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0]['content'] + '<|im_end|>\n' }}
+    {%- else %}
+        {{- '<|im_start|>system\nYou are Qwen, created by Alibaba Cloud. You are a helpful assistant.<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) or (message.role == "assistant" and not message.tool_calls) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {{- '<|im_start|>' + message.role }}
+        {%- if message.content %}
+            {{- '\n' + message.content }}
+        {%- endif %}
+        {%- for tool_call in message.tool_calls %}
+            {%- if tool_call.function is defined %}
+                {%- set tool_call = tool_call.function %}
+            {%- endif %}
+            {{- '\n<tool_call>\n{"name": "' }}
+            {{- tool_call.name }}
+            {{- '", "arguments": ' }}
+            {{- tool_call.arguments | tojson }}
+            {{- '}\n</tool_call>' }}
+        {%- endfor %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if (loop.index0 == 0) or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+{%- endif %}
--- a/config.json
+++ b/config.json
@@ -0,0 +1,57 @@
+{
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": null,
+  "dtype": "bfloat16",
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 896,
+  "initializer_range": 0.02,
+  "intermediate_size": 4864,
+  "layer_types": [
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention",
+    "full_attention"
+  ],
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "qwen2",
+  "num_attention_heads": 14,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 2,
+  "pad_token_id": 151643,
+  "rms_norm_eps": 1e-06,
+  "rope_parameters": {
+    "rope_theta": 1000000.0,
+    "rope_type": "default"
+  },
+  "sliding_window": null,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.12.1",
+  "use_cache": false,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}
--- a/generation_config.json
+++ b/generation_config.json
@@ -0,0 +1,13 @@
+{
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "temperature": 0.7,
+  "top_k": 20,
+  "top_p": 0.8,
+  "transformers_version": "5.12.1"
+}
--- a/model.safetensors
+++ b/model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7f21bccd6cd8bbef97fdb58907ee2cbccd1984bba1eb204b95a823d7ff8b42a
+size 988097824
--- a/tokenizer.json
+++ b/tokenizer.json
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7f96da3a872b5e901575b2067c744ad336c3a3d77a21584d20024557b1bd7f0
+size 11422059
--- a/tokenizer_config.json
+++ b/tokenizer_config.json
@@ -0,0 +1,30 @@
+{
+  "add_prefix_space": false,
+  "backend": "tokenizers",
+  "bos_token": null,
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "is_local": false,
+  "local_files_only": false,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}
--- a/training_args.bin
+++ b/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27eee94ef52ffdab13debdfd69595f23c360fa3f04a478919c1ce44e5a3d787b
+size 5777