arXiv subjects

Chengkun Yang

Publications and source records attributed to Chengkun Yang.

4 recordsLinked to original sources

arXiv2026

Towards Precision Therapy in Hepatocellular Carcinoma: A Clinical-Reasoning LLM for Risk Stratification and Treatment Guidance

Hepatocellular carcinoma (HCC) is a common malignancy and a leading cause of cancer-related mortality. Current guidelines and staging systems provide coarse categories, but often miss within-stage heterogeneity and the clinical context in electronic medical records (EMRs). We present HCC-STAR (Hepatocellular Carcinoma Staging, Treatment And pRognosis), a clinically aligned large language model that reads routine EMR narratives and jointly outputs risk score-based staging, ranked guideline-consistent treatments with evidence-based rationales, and individualized survival estimates. We curated about 30,000 HCC cases from SEER and expanded them into EMR-style narrative training data using a clinician-validated, prompt-based augmentation workflow. On this corpus, we developed a knowledge-aligned reasoning framework optimized with a step-verifiable composite reward, moving beyond text-level memorization of clinical guidelines. In a multi-center cohort of 6,668 patients from 12 hospitals in China, HCC-STAR achieved state-of-the-art performance in treatment recommendation and risk stratification compared with clinical guidelines and competitive models, including GPT-5 and Gemini-2.5 Pro. Hypothetical overall-survival analysis showed a median survival of 51 months under adherence to HCC-STAR recommendations, compared with 29 and 32 months under BCLC and CNLC. In clinician-centric evaluations, blinded hepatobiliary specialists rated HCC-STAR's reasoning and evidence-based justifications as trustworthy. The model surpassed resident and attending physicians in treatment accuracy and helped physicians make more accurate decisions faster when used as an assistant. These findings support HCC-STAR as a reliable and verifiable decision-support system for risk stratification and precision therapy in HCC.

cs.AI↗

arXiv2026

Clin-JEPA: A Multi-Phase Co-Training Framework for Joint-Embedding Predictive Pretraining on EHR Patient Trajectories

We present Clin-JEPA, a multi-phase co-training framework for joint-embedding predictive (JEPA) pretraining on EHR patient trajectories. JEPA architectures have enabled latent-space planning in robotics and high-quality representation learning in vision, but extending the paradigm to EHR data -- to obtain a single backbone that simultaneously forecasts patient trajectories and serves diverse downstream risk-prediction tasks without per-task fine-tuning -- remains an open challenge. Existing JEPA frameworks either discard the predictor after pretraining (I-JEPA, V-JEPA) or train it on a frozen pretrained encoder (V-JEPA 2-AC), leaving the encoder unaware of the rollout signal that the retained predictor must use at inference; co-training the encoder and predictor under a shared JEPA prediction objective would supply this grounding, but na\"ive co-training is unstable, with representation collapse and online/target drift causing autoregressive rollout to diverge. Clin-JEPA's five-phase pretraining curriculum -- predictor warmup, joint refinement, EMA target alignment, hard sync, and predictor finalization -- addresses each failure mode by phase, stably co-training a Qwen3-8B-based encoder and a 92M-parameter latent trajectory predictor. On MIMIC-IV ICU data, three independent evaluations support the framework: (1) latent $\ell_1$ rollout drift uniquely converges ($-$15.7%) over 48-hour horizons while baselines and ablations diverge (+3% to +4951%); (2) the encoder learns a clinically discriminative latent geometry (deteriorating-patient cohorts displace 4.83$\times$ further than stable patients in latent space, vs $\leq$2.62$\times$ for baseline encoders); (3) a single backbone outperforms strong tabular and sequence baselines on multi-task downstream evaluation. Clin-JEPA achieves mean AUROC 0.851 on ICareFM EEP and 0.883 on 8 binary risk tasks (+0.038 and +0.041 vs baseline average).

cs.LG↗

arXiv2025

Nonreciprocal optical circuit switching

Directly switching optical signals outperforms conventional optoelectronic hardware in terms of cost, latency, and energy efficiency, and is expected to address the growing demand for data node capacity driven by the development of machine learning and artificial intelligence (AI) technologies. Therefore, optical circuit switching (OCS) technology has piqued widespread research interest in various technical solutions, including silicon photonics. However, silicon-based integrated OCS remains constrained by challenges such as network performance and port scalability. Here we propose a magneto-optical heterogeneous integrated nonreciprocal OCS (NOCS) network based on a silicon photonics platform, achieving bidirectional full-duplex nonreciprocal transmission by programming reciprocal and nonreciprocal phase shifters. We demonstrate that compared with the existing OCS architecture, NOCS has the advantages of ultra-high reconfiguration speed, large-scale integration compatibility, and bidirectional channel isolation reducing the number of required ports. NOCS could meet the programming speed requirements of the AI backend network, or supports nonreciprocal optical switching applications without multiplexing technology.

physics.optics↗

arXiv2021

Burst-Error Propagation Suppression for Decision-Feedback Equalizer in Field-Trial Submarine Fiber-Optic Communications

In this paper, we present a field-trial C-band 72Gbit/s optical on-off keying (OOK) system over 18.8km dispersion-uncompensated submarine optical cable in the South China Sea. Chromatic dispersion (CD) of 18.8km submarine optical cable causes four spectral nulls on the 36GHz bandwidth of 72Gbit/s OOK signal, which is the main obstacle for achieving an acceptable bit-error-rate (BER) performance. Decision feedback equalizer (DFE) is effective to compensate for the spectral nulls. However, DFE has a serious defect of burst-error propagation when the burst errors emerge due to the unstable submarine environment. Weighted DFE (WDFE) can be used to mitigate the burst-error propagation, but it cannot fully compensate for the spectral nulls because only a part of feedback symbols is directly decided. Fortunately, maximum likelihood sequence estimation (MLSE) can be added after the WDFE to simultaneously eliminate the resisting spectral distortions and implement optimal detection. Compared to the joint DFE and MLSE algorithm, the joint WDFE and MLSE algorithm can effectively suppress the burst-error propagation to obtain a maximum 2.9dB improvement of $\boldsymbol{Q}$ factor and eliminate the phenomenon of BER floor. In conclusion, the joint WDFE and MLSE algorithm can solve the burst-error propagation for the field-trial fiber-optic communications.

cs.IT↗