Skip to content
Metatext Decide — typed decisions for your agents, with a probability on every answerGet early access
Metatext
Sign Up

Chinese-Jev

Paper introducing an encoder-only System One model for Chinese with domain fine-tunes and the CJ-Bench benchmark.

Paperresearch
Open source ↗

This summary is based on how curated lists describe the project; its README was not available to review.

Overview

Chinese-Jev is an arXiv paper presenting an encoder-only System One model for Chinese-language tasks, pretrained on 10 million examples and fine-tuned for medicine, law and finance, along with a benchmark called CJ-Bench.

Per the listing, the authors report higher general-domain accuracy than Jev with a 20x speedup.

What the decision model decides

TypeQuestionAnswers
ChoiceTyped classification of Chinese text in general, medical, legal and financial domains(inferred)—

How it works

  1. Pretrain an encoder on 10M examples
  2. Fine-tune for domains
  3. Evaluate on CJ-Bench against Jev

Strengths

  • Targets Chinese, where Jev is weaker
  • Domain variants

Caveats

  • Preprint; claims are the authors'
  • Only list summary available

At a glance

Type
Paper
Status
research

Use cases and integrations

How curated lists describe it

  • Encoder-only System One model for Chinese, pretrained on 10 million examples and fine-tuned for medicine, law, and finance, with the CJ-Bench benchmark; the authors report higher general-domain accuracy than Jev at a 20x speedup

    AnotiaWang/awesome-decision-models

Frequently asked questions

What is new?

An encoder-only decision model for Chinese with domain fine-tunes and CJ-Bench.

How does it compare to Jev?

The authors report higher general accuracy at 20x speed.

Where is it?

https://arxiv.org/abs/2609.36965

Related projects

  • TypeAR★ 946

    TypeLLM: Jev-inspired type-safe decoding for existing autoregressive LLMs on SGLang, with schema-guaranteed typed outputs.

    ProjectbetaPython
  • agentic-rl

    33-chapter Chinese tutorial on agentic RL and GRPO, including a chapter on when Jev-style external decision models beat training with RL.

    ProjectresearchPython
  • decision-model-audit

    Pre-registered field test of Jev, GLiNER2.5-Decide, Laya and other decision models against small LLMs on fresh September 2026 data.

    ProjectresearchPython
  • jev-dllm

    Reads Jev-style choice, yes/no and score probabilities from masked diffusion language models with shared Yes/No weights.

    ProjectresearchPython
  • metask-jev

    Open Jev-class decision models on Qwen3.5 with candidate-logit readout; the 4B model reports 80.1% on JevBench's 231 items.

    ProjectresearchPython
  • open_system_one

    Open System-1 (OpenJev): Apache-2.0 research models giving calibrated per-option probabilities, best is a 4.9B exp9a.

    ProjectresearchPython