Maya v0.1

LICENSE

@@ -0,0 +1,201 @@
+                                 Apache License
+                           Version 2.0, January 2004
+                        http://www.apache.org/licenses/
+
+   TERMS AND CONDITIONS FOR USE, REPRODUCTION, AND DISTRIBUTION
+
+   1. Definitions.
+
+      "License" shall mean the terms and conditions for use, reproduction,
+      and distribution as defined by Sections 1 through 9 of this document.
+
+      "Licensor" shall mean the copyright owner or entity authorized by
+      the copyright owner that is granting the License.
+
+      "Legal Entity" shall mean the union of the acting entity and all
+      other entities that control, are controlled by, or are under common
+      control with that entity. For the purposes of this definition,
+      "control" means (i) the power, direct or indirect, to cause the
+      direction or management of such entity, whether by contract or
+      otherwise, or (ii) ownership of fifty percent (50%) or more of the
+      outstanding shares, or (iii) beneficial ownership of such entity.
+
+      "You" (or "Your") shall mean an individual or Legal Entity
+      exercising permissions granted by this License.
+
+      "Source" form shall mean the preferred form for making modifications,
+      including but not limited to software source code, documentation
+      source, and configuration files.
+
+      "Object" form shall mean any form resulting from mechanical
+      transformation or translation of a Source form, including but
+      not limited to compiled object code, generated documentation,
+      and conversions to other media types.
+
+      "Work" shall mean the work of authorship, whether in Source or
+      Object form, made available under the License, as indicated by a
+      copyright notice that is included in or attached to the work
+      (an example is provided in the Appendix below).
+
+      "Derivative Works" shall mean any work, whether in Source or Object
+      form, that is based on (or derived from) the Work and for which the
+      editorial revisions, annotations, elaborations, or other modifications
+      represent, as a whole, an original work of authorship. For the purposes
+      of this License, Derivative Works shall not include works that remain
+      separable from, or merely link (or bind by name) to the interfaces of,
+      the Work and Derivative Works thereof.
+
+      "Contribution" shall mean any work of authorship, including
+      the original version of the Work and any modifications or additions
+      to that Work or Derivative Works thereof, that is intentionally
+      submitted to Licensor for inclusion in the Work by the copyright owner
+      or by an individual or Legal Entity authorized to submit on behalf of
+      the copyright owner. For the purposes of this definition, "submitted"
+      means any form of electronic, verbal, or written communication sent
+      to the Licensor or its representatives, including but not limited to
+      communication on electronic mailing lists, source code control systems,
+      and issue tracking systems that are managed by, or on behalf of, the
+      Licensor for the purpose of discussing and improving the Work, but
+      excluding communication that is conspicuously marked or otherwise
+      designated in writing by the copyright owner as "Not a Contribution."
+
+      "Contributor" shall mean Licensor and any individual or Legal Entity
+      on behalf of whom a Contribution has been received by Licensor and
+      subsequently incorporated within the Work.
+
+   2. Grant of Copyright License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      copyright license to reproduce, prepare Derivative Works of,
+      publicly display, publicly perform, sublicense, and distribute the
+      Work and such Derivative Works in Source or Object form.
+
+   3. Grant of Patent License. Subject to the terms and conditions of
+      this License, each Contributor hereby grants to You a perpetual,
+      worldwide, non-exclusive, no-charge, royalty-free, irrevocable
+      (except as stated in this section) patent license to make, have made,
+      use, offer to sell, sell, import, and otherwise transfer the Work,
+      where such license applies only to those patent claims licensable
+      by such Contributor that are necessarily infringed by their
+      Contribution(s) alone or by combination of their Contribution(s)
+      with the Work to which such Contribution(s) was submitted. If You
+      institute patent litigation against any entity (including a
+      cross-claim or counterclaim in a lawsuit) alleging that the Work
+      or a Contribution incorporated within the Work constitutes direct
+      or contributory patent infringement, then any patent licenses
+      granted to You under this License for that Work shall terminate
+      as of the date such litigation is filed.
+
+   4. Redistribution. You may reproduce and distribute copies of the
+      Work or Derivative Works thereof in any medium, with or without
+      modifications, and in Source or Object form, provided that You
+      meet the following conditions:
+
+      (a) You must give any other recipients of the Work or
+          Derivative Works a copy of this License; and
+
+      (b) You must cause any modified files to carry prominent notices
+          stating that You changed the files; and
+
+      (c) You must retain, in the Source form of any Derivative Works
+          that You distribute, all copyright, patent, trademark, and
+          attribution notices from the Source form of the Work,
+          excluding those notices that do not pertain to any part of
+          the Derivative Works; and
+
+      (d) If the Work includes a "NOTICE" text file as part of its
+          distribution, then any Derivative Works that You distribute must
+          include a readable copy of the attribution notices contained
+          within such NOTICE file, excluding those notices that do not
+          pertain to any part of the Derivative Works, in at least one
+          of the following places: within a NOTICE text file distributed
+          as part of the Derivative Works; within the Source form or
+          documentation, if provided along with the Derivative Works; or,
+          within a display generated by the Derivative Works, if and
+          wherever such third-party notices normally appear. The contents
+          of the NOTICE file are for informational purposes only and
+          do not modify the License. You may add Your own attribution
+          notices within Derivative Works that You distribute, alongside
+          or as an addendum to the NOTICE text from the Work, provided
+          that such additional attribution notices cannot be construed
+          as modifying the License.
+
+      You may add Your own copyright statement to Your modifications and
+      may provide additional or different license terms and conditions
+      for use, reproduction, or distribution of Your modifications, or
+      for any such Derivative Works as a whole, provided Your use,
+      reproduction, and distribution of the Work otherwise complies with
+      the conditions stated in this License.
+
+   5. Submission of Contributions. Unless You explicitly state otherwise,
+      any Contribution intentionally submitted for inclusion in the Work
+      by You to the Licensor shall be under the terms and conditions of
+      this License, without any additional terms or conditions.
+      Notwithstanding the above, nothing herein shall supersede or modify
+      the terms of any separate license agreement you may have executed
+      with Licensor regarding such Contributions.
+
+   6. Trademarks. This License does not grant permission to use the trade
+      names, trademarks, service marks, or product names of the Licensor,
+      except as required for reasonable and customary use in describing the
+      origin of the Work and reproducing the content of the NOTICE file.
+
+   7. Disclaimer of Warranty. Unless required by applicable law or
+      agreed to in writing, Licensor provides the Work (and each
+      Contributor provides its Contributions) on an "AS IS" BASIS,
+      WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or
+      implied, including, without limitation, any warranties or conditions
+      of TITLE, NON-INFRINGEMENT, MERCHANTABILITY, or FITNESS FOR A
+      PARTICULAR PURPOSE. You are solely responsible for determining the
+      appropriateness of using or redistributing the Work and assume any
+      risks associated with Your exercise of permissions under this License.
+
+   8. Limitation of Liability. In no event and under no legal theory,
+      whether in tort (including negligence), contract, or otherwise,
+      unless required by applicable law (such as deliberate and grossly
+      negligent acts) or agreed to in writing, shall any Contributor be
+      liable to You for damages, including any direct, indirect, special,
+      incidental, or consequential damages of any character arising as a
+      result of this License or out of the use or inability to use the
+      Work (including but not limited to damages for loss of goodwill,
+      work stoppage, computer failure or malfunction, or any and all
+      other commercial damages or losses), even if such Contributor
+      has been advised of the possibility of such damages.
+
+   9. Accepting Warranty or Additional Liability. While redistributing
+      the Work or Derivative Works thereof, You may choose to offer,
+      and charge a fee for, acceptance of support, warranty, indemnity,
+      or other liability obligations and/or rights consistent with this
+      License. However, in accepting such obligations, You may act only
+      on Your own behalf and on Your sole responsibility, not on behalf
+      of any other Contributor, and only if You agree to indemnify,
+      defend, and hold each Contributor harmless for any liability
+      incurred by, or claims asserted against, such Contributor by reason
+      of your accepting any such warranty or additional liability.
+
+   END OF TERMS AND CONDITIONS
+
+   APPENDIX: How to apply the Apache License to your work.
+
+      To apply the Apache License to your work, attach the following
+      boilerplate notice, with the fields enclosed by brackets "[]"
+      replaced with your own identifying information. (Don't include
+      the brackets!)  The text should be enclosed in the appropriate
+      comment syntax for the file format. We also recommend that a
+      file or class name and description of purpose be included on the
+      same "printed page" as the copyright notice for easier
+      identification within third-party archives.
+
+   Copyright [yyyy] [name of copyright owner]
+
+   Licensed under the Apache License, Version 2.0 (the "License");
+   you may not use this file except in compliance with the License.
+   You may obtain a copy of the License at
+
+       http://www.apache.org/licenses/LICENSE-2.0
+
+   Unless required by applicable law or agreed to in writing, software
+   distributed under the License is distributed on an "AS IS" BASIS,
+   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+   See the License for the specific language governing permissions and
+   limitations under the License.

NOTICE.md

@@ -0,0 +1,30 @@
+# Notice
+
+Maya is an unofficial experiment. It is not affiliated with or endorsed by ConvAI Innovations, Answer.AI, LightOn or the authors of the NLI models it is compared against.
+
+## This project
+
+The scripts, evaluation code and results in this repository are Copyright 2026 vishalmysore and licensed under the Apache License, Version 2.0 (see `LICENSE`). The evaluation cases in `data/eval` are copied from the author's layaForWeb repository (same license).
+
+`results/laya_moe_eval_reference.json` holds the laya-typed-decisions and layaMOE answers recorded by the author's layaMOE repository (https://github.com/vishalmysore/layaMOE).
+
+## Maya's weights
+
+Maya is fine-tuned from MoritzLaurer/ModernBERT-base-zeroshot-v2.0 (Apache-2.0, https://huggingface.co/MoritzLaurer/ModernBERT-base-zeroshot-v2.0), itself fine-tuned from answerdotai/ModernBERT-base (Apache-2.0). Changes: full fine-tune on Maya's synthetic yes/no data plus MNLI, labels renamed to yes/no. The released weights are a modified derivative.
+
+## Training data
+
+- `gen/laya_domains.py` is copied from the author's layaMOE repository (same license).
+- During training, a few thousand MNLI pairs (GLUE `mnli`, https://huggingface.co/datasets/nyu-mll/glue) are mixed in. They are downloaded at run time and not redistributed.
+
+## Models used as baselines
+
+Downloaded from Hugging Face at run time, not redistributed here. Check each model card for its license and training data before reusing a model.
+
+- convaiinnovations/laya-typed-decisions (Apache-2.0)
+- cross-encoder/nli-deberta-v3-xsmall
+- MoritzLaurer/deberta-v3-xsmall-zeroshot-v1.1-all-33
+- MoritzLaurer/deberta-v3-base-zeroshot-v2.0
+- MoritzLaurer/ModernBERT-base-zeroshot-v2.0
+- MoritzLaurer/deberta-v3-large-zeroshot-v2.0
+checkpoints/ (model weights) are not in git; they go to https://huggingface.co/VishalMysore/maya.

README.md

@@ -1,3 +1,40 @@
----
-license: apache-2.0
----
+---
+license: apache-2.0
+base_model: MoritzLaurer/ModernBERT-base-zeroshot-v2.0
+pipeline_tag: text-classification
+tags: [yes-no, guardrail, zero-shot, nli, modernbert, calibration]
+language: [en]
+---
+
+# Maya: a dedicated yes/no gate
+
+Maya reads a text and a statement (or a yes/no question) and returns P(yes). 150M parameters, fine-tuned from [MoritzLaurer/ModernBERT-base-zeroshot-v2.0](https://huggingface.co/MoritzLaurer/ModernBERT-base-zeroshot-v2.0) on rule-labeled synthetic data in which every fact appears in plain and negated wordings with opposite labels (plus a consistency loss that, in an ablation, added nothing measurable).
+
+Code, data, evaluation and full results: https://github.com/vishalmysore/maya. Browser (int8 ONNX) build: [VishalMysore/mayaWasm](https://huggingface.co/VishalMysore/mayaWasm).
+
+```python
+# pip install torch transformers ; then from the GitHub repo:
+from maya.gate import Maya
+maya = Maya.load("VishalMysore/maya")
+maya.ask("Hi team, how do I export the dashboard to Excel? No rush.", ["Is the customer angry?"])
+```
+
+Without the repo: tokenize the pair `(text, statement)`, run `AutoModelForSequenceClassification` (load with `dtype=torch.float32` on CPU), and take `sigmoid((logits[:, 0] - logits[:, 1]) / 2.8)`; label 0 is "yes". The temperature 2.8 is in `maya_config.json`.
+
+## Results (hand-labeled, never used for training)
+
+| | v1: 120 answers, mostly in-domain | v2: 256 answers, 8 unseen domains |
+|---|---|---|
+| laya-typed-decisions (421M) | 68.3% acc, AUROC 0.792 | 80.1% acc, AUROC 0.889 |
+| ModernBERT-base-zeroshot-v2.0 (starting point) | 76.7%, 0.875 | 78.5%, 0.868 |
+| deberta-v3-large-zeroshot-v2.0 (435M) | 75.8%, 0.890 | **89.8%, 0.957** |
+| **Maya** | **78.3%, 0.905** | 81.2%, 0.893 |
+
+Maya is the strongest model tested on its own domains (v1). On unseen domains (v2) it is level with Laya and smaller NLI models, and a 3x larger off-the-shelf NLI model is clearly better.
+
+- Contradictions between a statement and its negation: 25% (v1) and 27% (v2), against 44-92% for the other models.
+- With the shipped temperature, v2 calibration error (ECE) is 0.061.
+- Weak spots: agent guardrails (61% on v1, misled by lexical overlap such as "no human has reviewed" vs "without a human"), and minimal pairs in unseen domains (53% vs Laya's 69%). Do not use it as the only safety check.
+- It only outputs P(yes): every answer is yes or no (or "not sure" if you set abstention thresholds). It cannot refuse non-yes/no questions.
+
+Unofficial experiment, not affiliated with the authors of the base model or of Laya. Apache-2.0; see NOTICE.md.

config.json

@@ -0,0 +1,86 @@
+{
+  "architectures": [
+    "ModernBertForSequenceClassification"
+  ],
+  "attention_bias": false,
+  "attention_dropout": 0.0,
+  "bos_token_id": 50281,
+  "classifier_activation": "gelu",
+  "classifier_bias": false,
+  "classifier_dropout": 0.0,
+  "classifier_pooling": "mean",
+  "cls_token_id": 50281,
+  "decoder_bias": true,
+  "deterministic_flash_attn": false,
+  "dtype": "float32",
+  "embedding_dropout": 0.0,
+  "eos_token_id": 50282,
+  "global_attn_every_n_layers": 3,
+  "gradient_checkpointing": false,
+  "hidden_activation": "gelu",
+  "hidden_size": 768,
+  "id2label": {
+    "0": "yes",
+    "1": "no"
+  },
+  "initializer_cutoff_factor": 2.0,
+  "initializer_range": 0.02,
+  "intermediate_size": 1152,
+  "label2id": {
+    "no": 1,
+    "yes": 0
+  },
+  "layer_norm_eps": 1e-05,
+  "layer_types": [
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention",
+    "sliding_attention",
+    "sliding_attention",
+    "full_attention"
+  ],
+  "local_attention": 128,
+  "max_position_embeddings": 8192,
+  "mlp_bias": false,
+  "mlp_dropout": 0.0,
+  "model_type": "modernbert",
+  "norm_bias": false,
+  "norm_eps": 1e-05,
+  "num_attention_heads": 12,
+  "num_hidden_layers": 22,
+  "pad_token_id": 50283,
+  "position_embedding_type": "absolute",
+  "problem_type": "single_label_classification",
+  "rope_parameters": {
+    "full_attention": {
+      "rope_theta": 160000.0,
+      "rope_type": "default"
+    },
+    "sliding_attention": {
+      "rope_theta": 10000.0,
+      "rope_type": "default"
+    }
+  },
+  "sep_token_id": 50282,
+  "sparse_pred_ignore_index": -100,
+  "sparse_prediction": false,
+  "tie_word_embeddings": true,
+  "transformers_version": "5.17.0",
+  "vocab_size": 50368
+}

maya_config.json

@@ -0,0 +1,10 @@
+{
+ "p_yes": "sigmoid((logit_yes - logit_no) / temperature)",
+ "temperature": 2.8,
+ "temperature_fitted_on": "data/eval (v1, 120 hand-labeled answers); lowered v2 ECE from 0.152 to 0.061",
+ "thresholds": {
+  "t_yes": 0.5,
+  "t_no": 0.5
+ },
+ "thresholds_note": "Default is strict yes/no. For 'not sure' with a bounded error rate, fit maya.conformal.Abstainer on a labeled sample of your own inputs (a few hundred answers per side are needed to certify ~10% error)."
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cb7d67b8f8303f137bd70055ecdeeef542a3cd9c91d7c1599b15ee04a2bc6ea2
+size 598439784

tokenizer_config.json

@@ -0,0 +1,17 @@
+{
+  "backend": "tokenizers",
+  "clean_up_tokenization_spaces": true,
+  "cls_token": "[CLS]",
+  "is_local": false,
+  "local_files_only": false,
+  "mask_token": "[MASK]",
+  "model_input_names": [
+    "input_ids",
+    "attention_mask"
+  ],
+  "model_max_length": 512,
+  "pad_token": "[PAD]",
+  "sep_token": "[SEP]",
+  "tokenizer_class": "TokenizersBackend",
+  "unk_token": "[UNK]"
+}