CaoHaoWei

Jev-LCT-Qwen2.5-1.5B

README

License: apache-2.0

🌐 English Description

1. Key Innovations

In agentic workflows, API routing, and edge decision-making, conventional autoregressive LLMs suffer from high token-by-token generation latency and brittle string parsing, while small discriminative models produce systematically miscalibrated verbal confidence (overconfident hallucinations).

Jev-LCT (Looped Calibration Transformer) establishes a new paradigm for System-One Decision Models:

Parallel Looped Prefill: Recurrently iterates only the top $k = 2$ layers of Qwen2.5-1.5B with sequence right-shifting and Scale-Preserving RMS Injection, strictly preventing representation collapse.
Endogenous Trajectory Confidence: Extracts genuine calibrated confidence directly from hidden state convergence rates ( $Δ cos$ ), decision entropy reduction ( $Δ H$ ), and softmax margins—without requiring Reinforcement Learning (RLCD) or verbal introspection tokens.
- Error Detection AUROC reaches 0.9043 (In-Distribution) and 0.9524 (Out-of-Distribution).
- Expected Calibration Error (ECE-15) drops from 0.1646 to 0.1259 (a 23.5% relative reduction).
Dual-Channel Adaptive Early Exit:
- Confidence Margin Gate ( $m 1 \geq 0.90$ ): 85%+ of simple queries exit on loop 1 at single-pass latency (~45ms), completely eliminating overthinking.
- Cosine Attractor Stability ( $Δ t < 0.30$ ): Ambiguous problems dynamically iterate up to 4 loops before terminating cleanly.
- Deterministic SLA Guarantee: Keeps average latency at 61.9 ms on an RTX 3090 Ti.

2. Comprehensive 300-Item Benchmark (NVIDIA RTX 3090 Ti Verified)

Cross-domain benchmark comparison across financial intent (Banking77), scientific reasoning (AI2 ARC), factual verification (TruthfulQA), reading comprehension (BoolQ), and multi-task academic reasoning (MMLU):

Table with columns: Model Architecture, Parameters, Intent (Banking77), Science (ARC), Factuality (TQA), Reading (BoolQ), Academic (MMLU), Overall Accuracy, Avg Latency, Avg Loops
Model Architecture	Parameters	Intent (Banking77)	Science (ARC)	Factuality (TQA)	Reading (BoolQ)	Academic (MMLU)	Overall Accuracy	Avg Latency	Avg Loops
Convai Laya Base (ModernBERT)	421M	95.0%	28.3%	23.3%	76.7%	33.3%	51.3%

💡 Takeaway: At an ultra-fast 61.9 ms latency, Jev-LCT-1.5B surpasses Convai Laya by +41.7% and Open-Jev by +26.7% on scientific reasoning (ARC), delivering generative-grade reasoning at discriminative-grade speed.

3. Quickstart (3 Lines of Code)

This repository contains fully merged standalone weights (no separate adapter loading needed):

python

from lct_qwen_standalone import LCTQwen # 1. Load standalone model (bfloat16 on CUDA)model = LCTQwen.from_pretrained("CaoHaoWei/Jev-LCT-Qwen2.5-1.5B") # 2. Predict multiple-choice decisionresult = model.predict_choice(    prompt="Patient reports severe chest pain radiating to left shoulder. Determine triage urgency level:",    choices=["emergency", "urgent", "routine", "elective"]) print(f"Top Decision: {result['choice']}")print(f"Calibrated Confidence: {result['confidence']:.2%}")print(f"Loops Executed: {result['loops']}")

🇨🇳 中文详细说明

核心亮点

系统一决策新范式：结合 TypeSafe AI 的 Jev 理念，摆脱生成文本的高延迟，以毫秒级直出结构化决策与校准置信度。
通识与科学推理大幅领先：在 ARC 科学推理达到 71.7%（远超 Laya 30.0% 与 Open-Jev 45.0%），MMLU 达 58.3%。
免 RL 零额外开销校准：隐状态动力学余弦收敛与熵变轨迹特征，检错 AUROC 达 0.904 ~ 0.952，ECE 校准误差降低 23.5%。
自适应早停秒级直出：简单题单遍直出（~~45ms），复杂题深思 2~~4 轮，平均延迟仅 61.9 ms。

📜 Citation & License

Apache License 2.0. Please cite as:

bibtex

@article{cao2026lct,  title={Looped Calibration Transformer: Free Calibrated Confidence from Recurrent Computation Trajectories for Small Decision Models},  author={Cao, Haowei},  year={2026},  publisher={GitHub},  journal={GitHub repository},  howpublished={\url{https://github.com/gitchw/LCT}}}

Available on FriendliAI

Dedicated Endpoints

Run this model inference on single tenant GPU with unmatched speed and reliability at scale.

Learn more

Container

Run this model inference with full control and performance in your environment.

Learn more

Model Details

Model Provider

CaoHaoWei

Model Tree

Base

this model

Input Modalities

Text

Output Modalities