Datasets:
Tasks:
Text Classification
Formats:
parquet
Languages:
English
Size:
1K - 10K
Tags:
structured-decisions
calibration
probabilistic-classification
system-one
workflow-evaluation
Synthetic
License:
Submission: rahulkumar0612/lazydecider-0.8b (0.8B, specialist): accuracy 0.770 / KL 0.094 / Brier 0.051 / ECE 0.145
Browse files**Model:** [rahulkumar0612/lazydecider-0.8b](https://huggingface.co/rahulkumar0612/lazydecider-0.8b) (Qwen3.5-0.8B + Clef joint-decision head, LoRA fine-tuned)
**Mode:** specialist — fine-tuned on the `train` split for these four workflows.
**Scored on:** `test` (400 cases, 2,000 decisions), full probability distributions, using the benchmark's own `bench/score_with_jobs.py` arithmetic (reproduced verbatim in my scorer).
### Results (sealed `test`)
| Accuracy ↑ | KL from gold ↓ | Brier ↓ | ECE ↓ |
|---|---|---|---|
| 0.770 | **0.0938** | 0.0510 | 0.1454 |
The calibration column is the point: a **0.8B** model lands **KL 0.094** — near the top of the fitted/fine-tuned table, competitive with much larger models — while running fully on a laptop.
### Per-domain (specialist, per workflow)
| Workflow | Accuracy ↑ | KL ↓ |
|---|---|---|
| invoice_processing | 0.840 | 0.093 |
| customer_service | 0.796 | 0.115 |
| security_incidents | 0.738 | 0.074 |
| agent_trace_observability | 0.706 | 0.094 |
### How it was produced
The config was found **autonomously** by a local open-weight agent (Gemma 4 E4B) proposing hyperparameters overnight under an ASHA pruner, keeping a run only if it beat the current best on a frozen `select` split. The test set was **sealed** (SHA-256 committed; the search loop can't read it) and scored exactly once. A Tinker Qwen3.5-4B LoRA run independently confirmed the recipe lifts over the zero-shot baseline (0.601 vs 0.483).
- Code + full method: https://github.com/rahuls321/autoresearch-finetune-lab
- Live demo: https://lazydecider-demo.onrender.com
- The `.eval_results` YAML is already merged on the model repo (accuracy / kl_from_gold / brier / ece, config `all`, split `test`).
Happy to answer questions on the method or the sealing/scoring guarantees; raw `results.json` and per-domain scores are in the repo (`artifacts/test_score.json`, `artifacts/per_domain_scores.json`).
README.md
CHANGED
|
@@ -133,19 +133,20 @@ zero-shot and their scores are not comparable with the table above.
|
|
| 133 |
| 14 | [OpenDecider-medium-td](https://huggingface.co/manjunathshiva/opendecider-medium-td) (Qwen3-30B-A3B + LoRA) | general, then fine-tuned on `train` | 0.788 | 0.081 | 0.044 | – | 214 ms |
|
| 134 |
| 15 | [soft-decider-421m](https://huggingface.co/winwinwinbb/soft-decider-421m) | specialist | 0.774 | – | – | 0.141 | 50 ms |
|
| 135 |
| 16 | [Qwen3.5-4B-typed-decisions](https://huggingface.co/wozniakos10/Qwen3.5-4B-typed-decisions) (Qwen3.5-4B + LoRA, Clef head) | general, then fine-tuned on `train` | 0.7725 | 0.0992 | 0.0543 | 0.139 | – |
|
| 136 |
-
| 17 | [
|
| 137 |
-
| 18 | [
|
| 138 |
-
| 19 | [
|
| 139 |
-
| 20 | [
|
| 140 |
-
| 21 | [
|
| 141 |
-
| 22 | [
|
| 142 |
-
| 23 | [
|
| 143 |
-
| 24 | [
|
| 144 |
-
| 25 | [
|
| 145 |
-
| 26 |
|
| 146 |
-
| 27 |
|
| 147 |
-
| 28 |
|
| 148 |
-
| 29 | [Bekko System One v0
|
|
|
|
| 149 |
|
| 150 |
**Quyet-1.0-Large, served by Decision Models, has the highest accuracy among zero-shot models (0.791). [meraGPT Decider 1](https://meragpt.com/models/state-decider-1?utm_source=huggingface&utm_medium=dataset_card&utm_campaign=typed-decisions) is second on accuracy (0.768) and has the lowest KL (0.096) and Brier (0.052) of any zero-shot model.** It leads Jev on every question type (`noul` 0.840 vs
|
| 151 |
0.775, `choice` 0.733 vs 0.720, `score` 0.739 vs 0.696), its distributions sit far
|
|
|
|
| 133 |
| 14 | [OpenDecider-medium-td](https://huggingface.co/manjunathshiva/opendecider-medium-td) (Qwen3-30B-A3B + LoRA) | general, then fine-tuned on `train` | 0.788 | 0.081 | 0.044 | – | 214 ms |
|
| 134 |
| 15 | [soft-decider-421m](https://huggingface.co/winwinwinbb/soft-decider-421m) | specialist | 0.774 | – | – | 0.141 | 50 ms |
|
| 135 |
| 16 | [Qwen3.5-4B-typed-decisions](https://huggingface.co/wozniakos10/Qwen3.5-4B-typed-decisions) (Qwen3.5-4B + LoRA, Clef head) | general, then fine-tuned on `train` | 0.7725 | 0.0992 | 0.0543 | 0.139 | – |
|
| 136 |
+
| 17 | [lazydecider-0.8b](https://huggingface.co/rahulkumar0612/lazydecider-0.8b) (Qwen3.5-0.8B + Clef head, LoRA) ([#NN](https://huggingface.co/datasets/LocalLLaMA/typed-decisions/discussions/NN)) | fine-tuned on `train` (specialist, autonomous ASHA search) | 0.770 | 0.0938 | 0.0510 | 0.1454 | – |
|
| 137 |
+
| 18 | [embeddinggemma-2-ft-typed-decision](https://huggingface.co/mrm8488/embeddinggemma-2-ft-typed-decision) (0.3B) ([#37](https://huggingface.co/datasets/LocalLLaMA/typed-decisions/discussions/37)) | fine-tuned on `train` | 0.770 | 0.119 | 0.065 | 0.155 | – |
|
| 138 |
+
| 19 | [Laya typed-decisions](https://huggingface.co/convaiinnovations/laya-typed-decisions) | fitted on `train` | 0.766 | – | – | – | – |
|
| 139 |
+
| 20 | [gemma3-270m-typed-decisions](https://huggingface.co/rony/gemma3-270m-typed-decisions) (Gemma 3 270M, full fine-tune) | specialist | 0.761 | 0.132 | 0.072 | 0.148 | 68 ms |
|
| 140 |
+
| 21 | [SafeDecide-140M](https://huggingface.co/yotambr/safedecide-140m) (mmBERT-small, 140M) ([#39](https://huggingface.co/datasets/LocalLLaMA/typed-decisions/discussions/39)) | general, then fine-tuned on `train` | 0.754 | 0.126 | 0.069 | 0.147 | 80 ms (1 CPU thread) |
|
| 141 |
+
| 22 | [tasksource-decider-nano](https://huggingface.co/tasksource/tasksource-decider-nano) (149M cross-encoder, `v5`) ([#45](https://huggingface.co/datasets/LocalLLaMA/typed-decisions/discussions/45)) | general; `train` is 0.5% of its training mix | 0.7515 | 0.1125 | 0.0621 | 0.125 | 37 ms (A30, shared) |
|
| 142 |
+
| 23 | [ProKope-421M-8K](https://huggingface.co/ProKope-AI/ProKope-421M-8K) (421M, 8K context) | specialist | 0.7185 | – | 0.385 | – | 43 ms |
|
| 143 |
+
| 24 | [gemma3-270m-typed-decisions](https://huggingface.co/rony/gemma3-270m-typed-decisions) (Gemma 3 270M, full fine-tune) | general, fine-tuned on `train` | 0.702 | 0.156 | 0.091 | 0.128 | 68 ms |
|
| 144 |
+
| 25 | [Bekko System One v0 400M](https://huggingface.co/hotchpotch/bekko-system-one-v0-400m), scored by us | trained on `train` and other data | 0.668 | 0.214 | 0.113 | 0.143 | 808 ms |
|
| 145 |
+
| 26 | [vega-08b-public-intents](https://huggingface.co/nandakishorm/vega-08b-public-intents) (0.8B) ([#37](https://huggingface.co/datasets/LocalLLaMA/typed-decisions/discussions/37)) | specialist (bundled typed-decisions adapter active) | 0.663 | 0.435 | 0.191 | 0.074 | – |
|
| 146 |
+
| 27 | ModernBERT-base (149M), ours | specialist, fitted per workflow | 0.646 | 0.223 | 0.119 | 0.179 | 349 ms |
|
| 147 |
+
| 28 | MiniLM-L6 (22M), ours | specialist, fitted per workflow | 0.587 | 0.262 | 0.143 | 0.108 | 22 ms |
|
| 148 |
+
| 29 | [Bekko System One v0 68M](https://huggingface.co/hotchpotch/bekko-system-one-v0-68m), scored by us | trained on `train` and other data | 0.537 | 0.293 | 0.160 | 0.116 | 176 ms |
|
| 149 |
+
| 30 | [Bekko System One v0 17M](https://huggingface.co/hotchpotch/bekko-system-one-v0-17m), scored by us | trained on `train` and other data | 0.483 | 0.344 | 0.203 | 0.136 | 33 ms |
|
| 150 |
|
| 151 |
**Quyet-1.0-Large, served by Decision Models, has the highest accuracy among zero-shot models (0.791). [meraGPT Decider 1](https://meragpt.com/models/state-decider-1?utm_source=huggingface&utm_medium=dataset_card&utm_campaign=typed-decisions) is second on accuracy (0.768) and has the lowest KL (0.096) and Brier (0.052) of any zero-shot model.** It leads Jev on every question type (`noul` 0.840 vs
|
| 152 |
0.775, `choice` 0.733 vs 0.720, `score` 0.739 vs 0.696), its distributions sit far
|