South China Normal University

4 related articles

Subscribe to updates

researchsquare.com Oct 13, 2026
Breast cancer age-linked genes fall from 80 to four: only IL7R, SQSTM1, EGR1 and IGF1 survive replication and receptor robustness
Age-associated gene expression signals in breast tumors narrow from 80 to four candidates, IL7R, SQSTM1, EGR1 and IGF1, after receptor and stage adjustment, cross-platform replication and estrogen-receptor stratification. The same molecular workflow is also encoded as a bounded symbolic controller that completes, blocks or fails safe against a prespecified reference policy across 760 scenarios, rather than only finishing. The output is research prioritization, not a clinical biomarker or treatment recommendation.
Age-associated gene expression signals in breast tumors narrow from 80 to four candidates, IL7R, SQSTM1, EGR1 and IGF1, after receptor and stage adjustment, cross-platform replication and estrogen-receptor stratification. The same molecular workflow is also encoded as a bounded symbolic controller that completes, blocks or fails safe against a prespecified reference policy across 760 scenarios, rather than only finishing. The output is research prioritization, not a clinical biomarker or treatment recommendation.
South China Normal University Harbin Institute of Technology
Farah Deeba Memon（First author）
Bounded Agentic Orchestration Process-Aware Orchestration Benchmark Age-Aware Molecular Candidate Prioritization
arXiv Oct 12, 2026
Robots stop swallowing every experience whole: keeping only human-controlled segments of successful interventions lifts precision insertion success to 88.24%
Evo-RL filters deployment experience by estimated advantage before policy fine-tuning, raising attempt success rate on a real PiperX high-precision screw-insertion task from RECAP's 57.78% to 88.24%.
Central South University South China Normal University Shanghai Jiao Tong University
Kai Ye（First author）
Evo-RL Advantage-Guided Experience Selection RW-RL-Dataset Robot Reinforcement Learning
arXiv Oct 9, 2026
BayesJudge models conflicting human and LLM judgments as posterior distributions, raising expert prediction accuracy to 65.3% on SummEval
The authors propose BayesJudge, an online Bayesian meta-evaluation layer for conflicting human-LLM judgment streams that outputs a posterior verdict distribution per comparison while estimating rater-specific human confusion matrices and LLM presentation-order bias, using tie-open labels and paired AB/BA order swaps as protocol-level identification instruments; synthetic experiments recover prespecified evaluator parameters, and on SummEval it detects LLM position effects, separates expert from crowdworker behavior signatures without rater metadata, and yields posterior uncertainty that correlates with human disagreement.
The authors propose BayesJudge, an online Bayesian meta-evaluation layer for conflicting human-LLM judgment streams that outputs a posterior verdict distribution per comparison while estimating rater-specific human confusion matrices and LLM presentation-order bias, using tie-open labels and paired AB/BA order swaps as protocol-level identification instruments; synthetic experiments recover prespecified evaluator parameters, and on SummEval it detects LLM position effects, separates expert from crowdworker behavior signatures without rater metadata, and yields posterior uncertainty that correlates with human disagreement.
South China Normal University East China Normal University University of Southern California
Jiahao Zhang（First author）
Bayesian Meta-Evaluation LLM-as-a-Judge Position Bias Uncertainty Quantification
arXiv Sep 24, 2026
StableVQ: Turning Vector-Quantized Tokenizer Training Stability from a Fragile Outcome into a Guaranteed Property
The work proposes StableVQ, three parameter-free components — Dynamic STE, Region VQ Loss, and Decoupled Schedule — that disentangle Encoder–Decoder and Codebook training so each module can fulfill its own responsibility independently, consistently improving training stability, codebook utilization, and reconstruction quality on ImageNet across diverse codebook sizes and initialization settings.
The work proposes StableVQ, three parameter-free components — Dynamic STE, Region VQ Loss, and Decoupled Schedule — that disentangle Encoder–Decoder and Codebook training so each module can fulfill its own responsibility independently, consistently improving training stability, codebook utilization, and reconstruction quality on ImageNet across diverse codebook sizes and initialization settings.
Huazhong University of Science and Technology South China Normal University Klaipėda University
Bao Tang（First author）
StableVQ Vector Quantization Training Stability Codebook Utilization Dynamic Straight-Through Estimator

You’re all caught up

Breast cancer age-linked genes fall from 80 to four: only IL7R, SQSTM1, EGR1 and IGF1 survive replication and receptor robustness

Robots stop swallowing every experience whole: keeping only human-controlled segments of successful interventions lifts precision insertion success to 88.24%

BayesJudge models conflicting human and LLM judgments as posterior distributions, raising expert prediction accuracy to 65.3% on SummEval

StableVQ: Turning Vector-Quantized Tokenizer Training Stability from a Fragile Outcome into a Guaranteed Property