CountingSheep commited on
Commit
f9db65b
·
verified ·
1 Parent(s): 88387f0

Add the Qwen Apache-2.0 license, fix the loading examples, link the release; drop local_files_only

Browse files
Files changed (3) hide show
  1. LICENSE-Qwen +202 -0
  2. README.md +55 -22
  3. tokenizer_config.json +0 -1
LICENSE-Qwen ADDED
@@ -0,0 +1,202 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+
2
+ Apache License
3
+ Version 2.0, January 2004
4
+ http://www.apache.org/licenses/
5
+
6
+ TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
7
+
8
+ 1. Definitions.
9
+
10
+ "License" shall mean the terms and conditions for use, reproduction,
11
+ and distribution as defined by Sections 1 through 9 of this document.
12
+
13
+ "Licensor" shall mean the copyright owner or entity authorized by
14
+ the copyright owner that is granting the License.
15
+
16
+ "Legal Entity" shall mean the union of the acting entity and all
17
+ other entities that control, are controlled by, or are under common
18
+ control with that entity. For the purposes of this definition,
19
+ "control" means (i) the power, direct or indirect, to cause the
20
+ direction or management of such entity, whether by contract or
21
+ otherwise, or (ii) ownership of fifty percent (50%) or more of the
22
+ outstanding shares, or (iii) beneficial ownership of such entity.
23
+
24
+ "You" (or "Your") shall mean an individual or Legal Entity
25
+ exercising permissions granted by this License.
26
+
27
+ "Source" form shall mean the preferred form for making modifications,
28
+ including but not limited to software source code, documentation
29
+ source, and configuration files.
30
+
31
+ "Object" form shall mean any form resulting from mechanical
32
+ transformation or translation of a Source form, including but
33
+ not limited to compiled object code, generated documentation,
34
+ and conversions to other media types.
35
+
36
+ "Work" shall mean the work of authorship, whether in Source or
37
+ Object form, made available under the License, as indicated by a
38
+ copyright notice that is included in or attached to the work
39
+ (an example is provided in the Appendix below).
40
+
41
+ "Derivative Works" shall mean any work, whether in Source or Object
42
+ form, that is based on (or derived from) the Work and for which the
43
+ editorial revisions, annotations, elaborations, or other modifications
44
+ represent, as a whole, an original work of authorship. For the purposes
45
+ of this License, Derivative Works shall not include works that remain
46
+ separable from, or merely link (or bind by name) to the interfaces of,
47
+ the Work and Derivative Works thereof.
48
+
49
+ "Contribution" shall mean any work of authorship, including
50
+ the original version of the Work and any modifications or additions
51
+ to that Work or Derivative Works thereof, that is intentionally
52
+ submitted to Licensor for inclusion in the Work by the copyright owner
53
+ or by an individual or Legal Entity authorized to submit on behalf of
54
+ the copyright owner. For the purposes of this definition, "submitted"
55
+ means any form of electronic, verbal, or written communication sent
56
+ to the Licensor or its representatives, including but not limited to
57
+ communication on electronic mailing lists, source code control systems,
58
+ and issue tracking systems that are managed by, or on behalf of, the
59
+ Licensor for the purpose of discussing and improving the Work, but
60
+ excluding communication that is conspicuously marked or otherwise
61
+ designated in writing by the copyright owner as "Not a Contribution."
62
+
63
+ "Contributor" shall mean Licensor and any individual or Legal Entity
64
+ on behalf of whom a Contribution has been received by Licensor and
65
+ subsequently incorporated within the Work.
66
+
67
+ 2. Grant of Copyright License. Subject to the terms and conditions of
68
+ this License, each Contributor hereby grants to You a perpetual,
69
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
70
+ copyright license to reproduce, prepare Derivative Works of,
71
+ publicly display, publicly perform, sublicense, and distribute the
72
+ Work and such Derivative Works in Source or Object form.
73
+
74
+ 3. Grant of Patent License. Subject to the terms and conditions of
75
+ this License, each Contributor hereby grants to You a perpetual,
76
+ worldwide, non-exclusive, no-charge, royalty-free, irrevocable
77
+ (except as stated in this section) patent license to make, have made,
78
+ use, offer to sell, sell, import, and otherwise transfer the Work,
79
+ where such license applies only to those patent claims licensable
80
+ by such Contributor that are necessarily infringed by their
81
+ Contribution(s) alone or by combination of their Contribution(s)
82
+ with the Work to which such Contribution(s) was submitted. If You
83
+ institute patent litigation against any entity (including a
84
+ cross-claim or counterclaim in a lawsuit) alleging that the Work
85
+ or a Contribution incorporated within the Work constitutes direct
86
+ or contributory patent infringement, then any patent licenses
87
+ granted to You under this License for that Work shall terminate
88
+ as of the date such litigation is filed.
89
+
90
+ 4. Redistribution. You may reproduce and distribute copies of the
91
+ Work or Derivative Works thereof in any medium, with or without
92
+ modifications, and in Source or Object form, provided that You
93
+ meet the following conditions:
94
+
95
+ (a) You must give any other recipients of the Work or
96
+ Derivative Works a copy of this License; and
97
+
98
+ (b) You must cause any modified files to carry prominent notices
99
+ stating that You changed the files; and
100
+
101
+ (c) You must retain, in the Source form of any Derivative Works
102
+ that You distribute, all copyright, patent, trademark, and
103
+ attribution notices from the Source form of the Work,
104
+ excluding those notices that do not pertain to any part of
105
+ the Derivative Works; and
106
+
107
+ (d) If the Work includes a "NOTICE" text file as part of its
108
+ distribution, then any Derivative Works that You distribute must
109
+ include a readable copy of the attribution notices contained
110
+ within such NOTICE file, excluding those notices that do not
111
+ pertain to any part of the Derivative Works, in at least one
112
+ of the following places: within a NOTICE text file distributed
113
+ as part of the Derivative Works; within the Source form or
114
+ documentation, if provided along with the Derivative Works; or,
115
+ within a display generated by the Derivative Works, if and
116
+ wherever such third-party notices normally appear. The contents
117
+ of the NOTICE file are for informational purposes only and
118
+ do not modify the License. You may add Your own attribution
119
+ notices within Derivative Works that You distribute, alongside
120
+ or as an addendum to the NOTICE text from the Work, provided
121
+ that such additional attribution notices cannot be construed
122
+ as modifying the License.
123
+
124
+ You may add Your own copyright statement to Your modifications and
125
+ may provide additional or different license terms and conditions
126
+ for use, reproduction, or distribution of Your modifications, or
127
+ for any such Derivative Works as a whole, provided Your use,
128
+ reproduction, and distribution of the Work otherwise complies with
129
+ the conditions stated in this License.
130
+
131
+ 5. Submission of Contributions. Unless You explicitly state otherwise,
132
+ any Contribution intentionally submitted for inclusion in the Work
133
+ by You to the Licensor shall be under the terms and conditions of
134
+ this License, without any additional terms or conditions.
135
+ Notwithstanding the above, nothing herein shall supersede or modify
136
+ the terms of any separate license agreement you may have executed
137
+ with Licensor regarding such Contributions.
138
+
139
+ 6. Trademarks. This License does not grant permission to use the trade
140
+ names, trademarks, service marks, or product names of the Licensor,
141
+ except as required for reasonable and customary use in describing the
142
+ origin of the Work and reproducing the content of the NOTICE file.
143
+
144
+ 7. Disclaimer of Warranty. Unless required by applicable law or
145
+ agreed to in writing, Licensor provides the Work (and each
146
+ Contributor provides its Contributions) on an "AS IS" BASIS,
147
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
148
+ implied, including, without limitation, any warranties or conditions
149
+ of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
150
+ PARTICULAR PURPOSE. You are solely responsible for determining the
151
+ appropriateness of using or redistributing the Work and assume any
152
+ risks associated with Your exercise of permissions under this License.
153
+
154
+ 8. Limitation of Liability. In no event and under no legal theory,
155
+ whether in tort (including negligence), contract, or otherwise,
156
+ unless required by applicable law (such as deliberate and grossly
157
+ negligent acts) or agreed to in writing, shall any Contributor be
158
+ liable to You for damages, including any direct, indirect, special,
159
+ incidental, or consequential damages of any character arising as a
160
+ result of this License or out of the use or inability to use the
161
+ Work (including but not limited to damages for loss of goodwill,
162
+ work stoppage, computer failure or malfunction, or any and all
163
+ other commercial damages or losses), even if such Contributor
164
+ has been advised of the possibility of such damages.
165
+
166
+ 9. Accepting Warranty or Additional Liability. While redistributing
167
+ the Work or Derivative Works thereof, You may choose to offer,
168
+ and charge a fee for, acceptance of support, warranty, indemnity,
169
+ or other liability obligations and/or rights consistent with this
170
+ License. However, in accepting such obligations, You may act only
171
+ on Your own behalf and on Your sole responsibility, not on behalf
172
+ of any other Contributor, and only if You agree to indemnify,
173
+ defend, and hold each Contributor harmless for any liability
174
+ incurred by, or claims asserted against, such Contributor by reason
175
+ of your accepting any such warranty or additional liability.
176
+
177
+ END OF TERMS AND CONDITIONS
178
+
179
+ APPENDIX: How to apply the Apache License to your work.
180
+
181
+ To apply the Apache License to your work, attach the following
182
+ boilerplate notice, with the fields enclosed by brackets "[]"
183
+ replaced with your own identifying information. (Don't include
184
+ the brackets!) The text should be enclosed in the appropriate
185
+ comment syntax for the file format. We also recommend that a
186
+ file or class name and description of purpose be included on the
187
+ same "printed page" as the copyright notice for easier
188
+ identification within third-party archives.
189
+
190
+ Copyright 2026 Alibaba Cloud
191
+
192
+ Licensed under the Apache License, Version 2.0 (the "License");
193
+ you may not use this file except in compliance with the License.
194
+ You may obtain a copy of the License at
195
+
196
+ http://www.apache.org/licenses/LICENSE-2.0
197
+
198
+ Unless required by applicable law or agreed to in writing, software
199
+ distributed under the License is distributed on an "AS IS" BASIS,
200
+ WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
201
+ See the License for the specific language governing permissions and
202
+ limitations under the License.
README.md CHANGED
@@ -5,6 +5,7 @@ base_model_relation: finetune
5
  library_name: transformers
6
  pipeline_tag: image-text-to-text
7
  language: [en, zh]
 
8
  tags: [judgment, classification, calibration, vision-language, system-one, vev]
9
  model-index:
10
  - name: vev-4b
@@ -36,19 +37,24 @@ model-index:
36
 
37
  Vev is a judgment model with open weights. Given a state (text, JSON, images, or a mix) and typed questions
38
  (yes/no, multiple choice, graded score), it returns a probability for every allowed answer in one forward pass,
39
- without generating text. `vev-4b` is a LoRA fine-tune of [Qwen/Qwen3.5-4B](https://huggingface.co/Qwen/Qwen3.5-4B).
40
- The `vev` server exposes it through an HTTP API with the same request and response format as TypeSafe's
41
- `/v1/systemone`.
42
-
43
- This is v0.1, a research preview, released for non-commercial use under CC BY-NC 4.0.
44
-
45
- This repository holds merged weights: the adapter is already folded into the base model. The adapter alone is in
46
- [CountingSheep/vev-4b-lora](https://huggingface.co/CountingSheep/vev-4b-lora).
 
 
 
 
 
47
 
48
  ## Usage
49
 
50
  ```bash
51
- pip install git+https://github.com/Xiaooolong/vev
52
  vev serve --model CountingSheep/vev-4b
53
  ```
54
 
@@ -67,18 +73,42 @@ resp = client.system_one(
67
  print(resp.answers["urgent"].noul, resp.answers["team"].probabilities)
68
  ```
69
 
70
- Images go into the state as `{"image": {"url": "data:image/png;base64,..."}}`. The API specification, the server
71
- and the evaluation code are at [https://github.com/Xiaooolong/vev](https://github.com/Xiaooolong/vev).
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
72
 
73
- The weights also load with plain `transformers` as a Qwen3.5 vision-language model. Vev reads its answers from the
74
- next-token probabilities of the answer tokens (Yes/No, option letters, level digits) at the end of a fixed prompt;
75
- the prompt is built in `vev/model.py` in the code repository.
76
 
77
  ## Evaluation
78
 
79
  Accuracy on full sets with the harness in the code repository. The zero-shot column is the base model read the same
80
- way without fine-tuning. kev-4B was run locally with the same harness; Jev was queried through TypeSafe's API
81
- (`jev-latest`, 23–24 September 2026); no Jev outputs were used for training.
 
 
 
 
82
 
83
  | Set | n | Qwen3.5-4B zero-shot | vev-4b | kev-4B | Jev |
84
  |---|---|---|---|---|---|
@@ -101,15 +131,16 @@ way without fine-tuning. kev-4B was run locally with the same harness; Jev was q
101
 
102
  The last six rows are judgment sets we converted from human-labelled public data; they are not established
103
  benchmarks. Brier score and calibration error for every set are in `results/` in the code repository; significance
104
- tests against the base model, kev-4B and Jev are marked in its README.
105
 
106
  Reversing the order of the options changes the top answer on 17% of JevBench and kev transfer-v4 questions (kev-4B:
107
  10–12%, Jev: under 4%).
108
 
109
  ## Limitations
110
 
111
- - Jev is more accurate than Vev on nimble and kev transfer-v4 (by 18–22 and 7–8 points), and than vev-4b on
112
- JevBench.
 
113
  - Answers depend on option order more than kev-4B's and Jev's do (rates above).
114
  - Judgments that need several reasoning steps are weaker than the base model's own answer when it may think first
115
  (4–8 points on one internal scenario set).
@@ -120,14 +151,16 @@ Reversing the order of the options changes the top answer on 17% of JevBench and
120
 
121
  LoRA rank 16 on all linear layers of the language model, vision tower frozen; 2,500 steps on 100,000 records
122
  sampled from a mixture of 42 public text and image datasets (English and Chinese), with a KL term that keeps the
123
- answer distribution close to the base model's on training rows and on 10,123 extra unlabeled anchor rows. One
124
- NVIDIA H800. The exact recipe, library versions and the list of sources with their licenses are in
 
125
  [TRAINING.md](https://github.com/Xiaooolong/vev/blob/main/TRAINING.md).
126
 
127
  ## License
128
 
129
  CC BY-NC 4.0. Six of the training sources are research-only or non-commercial and eleven have no clear license for
130
- their data; this license does not replace their terms. The base model, Qwen/Qwen3.5-4B, is Apache 2.0.
 
131
 
132
  ## Citation
133
 
 
5
  library_name: transformers
6
  pipeline_tag: image-text-to-text
7
  language: [en, zh]
8
+ datasets: [CountingSheep/vev-anchor-v3]
9
  tags: [judgment, classification, calibration, vision-language, system-one, vev]
10
  model-index:
11
  - name: vev-4b
 
37
 
38
  Vev is a judgment model with open weights. Given a state (text, JSON, images, or a mix) and typed questions
39
  (yes/no, multiple choice, graded score), it returns a probability for every allowed answer in one forward pass,
40
+ without generating text. `vev-4b` is a LoRA fine-tune of [Qwen/Qwen3.5-4B](https://huggingface.co/Qwen/Qwen3.5-4B),
41
+ with the adapter merged into the weights. The adapter alone is in
42
+ [CountingSheep/vev-4b-lora](https://huggingface.co/CountingSheep/vev-4b-lora). The `vev` server exposes it through
43
+ an HTTP API with the same request and response format as TypeSafe's `/v1/systemone`.
44
+
45
+ This is v0.1, a research preview, released for non-commercial use under CC BY-NC 4.0. Code:
46
+ [Xiaooolong/vev](https://github.com/Xiaooolong/vev). All models of this release:
47
+ [vev-4b](https://huggingface.co/CountingSheep/vev-4b) ·
48
+ [vev-4b-lora](https://huggingface.co/CountingSheep/vev-4b-lora) ·
49
+ [vev-9b](https://huggingface.co/CountingSheep/vev-9b) ·
50
+ [vev-9b-lora](https://huggingface.co/CountingSheep/vev-9b-lora); training anchor data:
51
+ [CountingSheep/vev-anchor-v3](https://huggingface.co/datasets/CountingSheep/vev-anchor-v3);
52
+ [collection](https://huggingface.co/collections/CountingSheep/vev-v01-6abd3e17303828e10c49dc65).
53
 
54
  ## Usage
55
 
56
  ```bash
57
+ pip install git+https://github.com/Xiaooolong/vev typesafe-sdk
58
  vev serve --model CountingSheep/vev-4b
59
  ```
60
 
 
73
  print(resp.answers["urgent"].noul, resp.answers["team"].probabilities)
74
  ```
75
 
76
+ `Noul` is the API's name for a yes/no question; `.noul` is the probability of "yes". Images go into the state as
77
+ `{"image": {"url": "data:image/png;base64,..."}}`. The API is specified in
78
+ [spec/systemone-api.md](https://github.com/Xiaooolong/vev/blob/main/spec/systemone-api.md).
79
+
80
+ To pin this version, download the tagged revision and serve the local copy:
81
+
82
+ ```bash
83
+ hf download CountingSheep/vev-4b --revision v0.1.0 --local-dir vev-4b-v0.1.0
84
+ vev serve --model vev-4b-v0.1.0
85
+ ```
86
+
87
+ ### Loading without the server
88
+
89
+ ```python
90
+ from transformers import AutoModelForImageTextToText, AutoProcessor
91
+
92
+ model = AutoModelForImageTextToText.from_pretrained("CountingSheep/vev-4b", dtype="bfloat16")
93
+ processor = AutoProcessor.from_pretrained("CountingSheep/vev-4b")
94
+ ```
95
+
96
+ `config.json` keeps `mtp_num_hidden_layers` from the base model, but the multi-token-prediction weights are not
97
+ included; reading answers does not use them.
98
 
99
+ This gives the language model only. Vev reads its answers from the next-token probabilities of the answer tokens
100
+ (Yes/No, option letters, level digits) at the end of a fixed prompt, which is built in
101
+ [vev/model.py](https://github.com/Xiaooolong/vev/blob/main/vev/model.py).
102
 
103
  ## Evaluation
104
 
105
  Accuracy on full sets with the harness in the code repository. The zero-shot column is the base model read the same
106
+ way without fine-tuning. [Jev](https://typesafe.ai/blog/introducing-system-one-models-and-jev) is TypeSafe's hosted
107
+ judgment model, queried through its API (`jev-latest`, 23–24 September 2026); no Jev outputs were used for
108
+ training. [kev-4B](https://github.com/jaredpalmer/kev) is an open judgment model on a 4B Qwen base, run locally
109
+ with the same harness. The text sets are [judgekit](https://github.com/lexingtonhibiki/judgekit) (Chinese),
110
+ [JevBench](https://github.com/fstandhartinger/jevbench) (public subset),
111
+ [nimble](https://github.com/bespokelabsai/nimble) and kev's own held-out test set, transfer-v4.
112
 
113
  | Set | n | Qwen3.5-4B zero-shot | vev-4b | kev-4B | Jev |
114
  |---|---|---|---|---|---|
 
131
 
132
  The last six rows are judgment sets we converted from human-labelled public data; they are not established
133
  benchmarks. Brier score and calibration error for every set are in `results/` in the code repository; significance
134
+ tests against the base model, kev-4B and Jev are in its README.
135
 
136
  Reversing the order of the options changes the top answer on 17% of JevBench and kev transfer-v4 questions (kev-4B:
137
  10–12%, Jev: under 4%).
138
 
139
  ## Limitations
140
 
141
+ - Jev is significantly more accurate on nimble, kev transfer-v4 and JevBench (by 22, 8 and 10 points).
142
+ - On POPE, vev-4b is 0.5 points less accurate than Qwen3.5-4B and its Brier score is 0.007 worse; both differences
143
+ are significant. It is the only set where fine-tuning made the 4B model significantly worse.
144
  - Answers depend on option order more than kev-4B's and Jev's do (rates above).
145
  - Judgments that need several reasoning steps are weaker than the base model's own answer when it may think first
146
  (4–8 points on one internal scenario set).
 
151
 
152
  LoRA rank 16 on all linear layers of the language model, vision tower frozen; 2,500 steps on 100,000 records
153
  sampled from a mixture of 42 public text and image datasets (English and Chinese), with a KL term that keeps the
154
+ answer distribution close to the base model's on training rows and on the 10,123 unlabeled rows of
155
+ [CountingSheep/vev-anchor-v3](https://huggingface.co/datasets/CountingSheep/vev-anchor-v3). One NVIDIA H800. The
156
+ exact recipe, library versions and the list of sources with their licenses are in
157
  [TRAINING.md](https://github.com/Xiaooolong/vev/blob/main/TRAINING.md).
158
 
159
  ## License
160
 
161
  CC BY-NC 4.0. Six of the training sources are research-only or non-commercial and eleven have no clear license for
162
+ their data; this license does not replace their terms. The base model, Qwen/Qwen3.5-4B, is licensed under the
163
+ Apache License 2.0, included here as `LICENSE-Qwen`.
164
 
165
  ## Citation
166
 
tokenizer_config.json CHANGED
@@ -10,7 +10,6 @@
10
  "errors": "replace",
11
  "image_token": "<|image_pad|>",
12
  "is_local": false,
13
- "local_files_only": true,
14
  "model_max_length": 262144,
15
  "model_specific_special_tokens": {
16
  "audio_bos_token": "<|audio_start|>",
 
10
  "errors": "replace",
11
  "image_token": "<|image_pad|>",
12
  "is_local": false,
 
13
  "model_max_length": 262144,
14
  "model_specific_special_tokens": {
15
  "audio_bos_token": "<|audio_start|>",