Non-autoregressive System 1 decision engine. Typed choice, score and yes/no decisions over any text in a single forward pass, in 100+ languages, with a router that picks the right checkpoint per request.
Python calibrationclassificationdecision-modelhuggingfacejevmodernbertmultilingualnlppythonpytorchroutingtyped-decisionszero-shot
Merge pull request #389 from PerryLink/research-refresh-51-language-sweep
Nandakishor14379cbf7d66
diff --cc research/README.mdindex 2fbb00d1,4e879b54..c6003b03--- a/research/README.md+++ b/research/README.md@@@ -104,10 -77,9 +104,11 @@@ python research/scripts/bench_length_ba|---|---|| `results/t4_colab_benchmark.json` | 17,416 questions on one T4, both checkpoints, identical questions per model || `results/long_context_multilingual.json` | the long-document run behind `assets/long_context_8192.png`: every prediction, with device and library versions |-| `results/cpu_51_language_sweep.json` | 51 languages x 2 checkpoints, MASSIVE intent, 20 options |-| `results/cpu_51_language_sweep_clamped.json` | the English half of the 51 languages re-run after the temperature clamp, raw temperatures and served temperatures side by side ([#222](https://github.com/NandhaKishorM/laya/pull/222)) |+| `results/cpu_51_language_sweep.json` | 51 languages x 2 checkpoints, MASSIVE intent, 20 options (committed `part_b` covers the English checkpoint only; a rerun of `research/scripts/bench_local.py` writes the gitignored `research/local_benchmark_results.json` and leaves this file untouched) |+| `results/cpu_51_language_sweep_clamped.json` | the same 51 languages and 5,100 cases re-run after the temperature clamp, raw temperatures and served temperatures side by side ([#208](https://github.com/NandhaKishorM/laya/issues/208)) |+ | `results/cpu_51_language_sweep_refreshed.json` | both checkpoints with the environment recorded, the superseded committed columns kept beside the new ones. The multilingual half of the committed sweep does not reproduce on current code, so this is what the table in `BENCHMARKS.md` prints ([#208](https://github.com/NandhaKishorM/laya/issues/208)) |+| `results/app_benchmark_results.json` | Applications run (400 cases per task, seed 13, 2026-09-19); source for the BENCHMARKS Themes / Jev-dataset tables and the README routed AG News / DAIR Emotion cells (a rerun of `research/scripts/bench_apps.py` writes the gitignored `research/app_benchmark_results.json` instead) |+| `research/latency_benchmark_results.json` | **not committed** (gitignored) — generate with `research/scripts/bench_latency.py`; CPU routing-cost regimes behind the README preload table |## Headline findings