City University of Hong Kong

15 related articles

Subscribe to updates

arXiv Oct 27, 2026
Reconstruction residuals miss a drifting sensor: CLOSURE adds a second, separately learned channel-consistency score and reaches 88.13 Aff-F1 on SMD
Multivariate time series anomaly detection has long leaned on reconstruction residuals, but a residual only certifies that a window can be reproduced, not that its channels still explain one another as they do under normal operation. CLOSURE splits recoverability and cross-channel consistency into two prediction tasks trained separately on normal data: a diffusion reconstructor conditioned on a discrete memory of normal temporal bases scores the first, and a relation operator that predicts each channel only from its peers scores the second, with the two combined after calibration on normal data. Across the five public benchmarks SMD, MSL, SMAP, SWaT and PSM, its Aff-F1 exceeds the listed methods, reaching 88.13 on SMD.
Multivariate time series anomaly detection has long leaned on reconstruction residuals, but a residual only certifies that a window can be reproduced, not that its channels still explain one another as they do under normal operation. CLOSURE splits recoverability and cross-channel consistency into two prediction tasks trained separately on normal data: a diffusion reconstructor conditioned on a discrete memory of normal temporal bases scores the first, and a relation operator that predicts each channel only from its peers scores the second, with the two combined after calibration on normal data. Across the five public benchmarks SMD, MSL, SMAP, SWaT and PSM, its Aff-F1 exceeds the listed methods, reaching 88.13 on SMD.
Tianjin University Beijing Jiaotong University Shanghai Jiao Tong University Tsinghua University City University of Hong Kong
Fan Xu（First author）
CLOSURE Time-Series Anomaly Detection Relation-Closure Operator Diffusion Model
bioRxiv Nov 9, 2026
TriStageHLA splits epitope discovery into three stages, lifting immunogenicity AUROC from 60.1% to 81.4%
A team including City University of Hong Kong released TriStageHLA, which splits HLA class I epitope discovery into peptide-only screening, allele-specific binding prediction and immunogenicity prediction in sequence. On held-out immunogenicity data the immunogenicity module reached AUROC 81.4% and AUPRC 73.8%, above ImmuneApp at 70.7%/63.6% and BigMHC-IM at 60.1%/51.5%; the binding module reached AUROC 96.2% and AUPRC 95.2% on combined BA/EL evaluation. In a simulated screen over 300,000 human-proteome background peptides, the screening stage kept 82.4% of true candidate peptides while discarding 87% of background peptides.
A team including City University of Hong Kong released TriStageHLA, which splits HLA class I epitope discovery into peptide-only screening, allele-specific binding prediction and immunogenicity prediction in sequence. On held-out immunogenicity data the immunogenicity module reached AUROC 81.4% and AUPRC 73.8%, above ImmuneApp at 70.7%/63.6% and BigMHC-IM at 60.1%/51.5%; the binding module reached AUROC 96.2% and AUPRC 95.2% on combined BA/EL evaluation. In a simulated screen over 300,000 human-proteome background peptides, the screening stage kept 82.4% of true candidate peptides while discarding 87% of background peptides.
Jilin University Lanzhou University City University of Hong Kong Harvard University
Tianchi Lu（First author）Ka‐Chun Wong（Corresponding author）Xiangtao Li（Corresponding author）
TriStageHLA peptide–HLA binding prediction Immunogenicity Prediction Low-Rank Adaptation
arXiv Oct 25, 2026
Tail calibration flags when a larger Best-of-N budget backfires, and code generation is where it shows
Best-of-N inference returns the candidate with the highest proxy reward, so a larger budget reads the extreme tail of the reward distribution rather than its average ranking quality. The work writes the BoN value as a Beta-weighted integral of conditional true reward over within-prompt score quantiles, showing that BoN depends on scores only through their ranks and can decline only when the calibration curve in the tail turns downward. On public candidate pools, negative tail excess concentrates in code generation, and a cross-fitted tail statistic predicts whether a budget cap pays off on held-out data about as well as rank accuracy.
Best-of-N inference returns the candidate with the highest proxy reward, so a larger budget reads the extreme tail of the reward distribution rather than its average ranking quality. The work writes the BoN value as a Beta-weighted integral of conditional true reward over within-prompt score quantiles, showing that BoN depends on scores only through their ranks and can decline only when the calibration curve in the tail turns downward. On public candidate pools, negative tail excess concentrates in code generation, and a cross-fitted tail statistic predicts whether a budget cap pays off on held-out data about as well as rank accuracy.
Hong Kong Polytechnic University University of Washington University of Science and Technology Beijing City University of Hong Kong
Cheng Gui（First author）
Reward Hacking Tail Calibration Reward Model
medRxiv Oct 5, 2026
Colorectal cancer recurrence after surgery gets a spatial host-microbiome readout: AUC 0.81 versus 0.45 for standard pathology
Whether colorectal cancer recurs after resection has been judged mainly from routine pathology such as lymphovascular invasion, perineural invasion, tumor budding and Ki-67. AlphaFISH pairs a sequencing-free nanoprobe biochip with a transformer model to read 41 host transcripts and 51 bacterial genera across 149 colorectal biopsies, and in 28 patients followed for one year after surgery it predicted recurrence or metastasis with an AUC of 0.81, against 0.45 for the combined routine clinical markers. The result comes from a preprint that has not been peer reviewed, on a limited sample.
Whether colorectal cancer recurs after resection has been judged mainly from routine pathology such as lymphovascular invasion, perineural invasion, tumor budding and Ki-67. AlphaFISH pairs a sequencing-free nanoprobe biochip with a transformer model to read 41 host transcripts and 51 bacterial genera across 149 colorectal biopsies, and in 28 patients followed for one year after surgery it predicted recurrence or metastasis with an AUC of 0.81, against 0.45 for the combined routine clinical markers. The result comes from a preprint that has not been peer reviewed, on a limited sample.
Sixth Affiliated Hospital of Sun Yat-sen University Sun Yat-sen University City Of Hope National Medical Center University of Hong Kong Tsinghua University City University of Hong Kong
Chenxi Hu（First author）Hailiang Sun（First author）Feng Guo（First author）Jianzhu Ma（Corresponding author）Peng Shi（Corresponding author）
AlphaFISH Spatial transcriptomics Transformer microbiome colorectal cancer
researchsquare.com Oct 22, 2026
A tactile skin that needs just one wire: super-resolution factor above 10,000 from palm-sized to back-sized patches
Large-area tactile sensing has meant tiling many sensing units and routing a wire to each, so scaling area or resolution drives up wiring and cost. A serpentine electromagnetic surface-wave transmission line on flexible polyimide is fed swept signals from both ends, and the dual-port reflection spectra encode contact position, which a multi-branch multilayer perceptron decodes into coordinates. Two prototypes an order of magnitude apart in area, 5×5 and 16×18 cm, both exceed a super-resolution factor of 10,000, with two-dimensional localization RMSE of 0.368 mm for the small skin and 1.209 mm for the large one.
Large-area tactile sensing has meant tiling many sensing units and routing a wire to each, so scaling area or resolution drives up wiring and cost. A serpentine electromagnetic surface-wave transmission line on flexible polyimide is fed swept signals from both ends, and the dual-port reflection spectra encode contact position, which a multi-branch multilayer perceptron decodes into coordinates. Two prototypes an order of magnitude apart in area, 5×5 and 16×18 cm, both exceed a super-resolution factor of 10,000, with two-dimensional localization RMSE of 0.368 mm for the small skin and 1.209 mm for the large one.
City University of Hong Kong
Arindam Basu（First author）
Multi-Branch Multilayer Perceptron Electromagnetic Skin Tactile Sensing Gradient-Boosted Trees (XGBoost)
arXiv Oct 12, 2026
Agent workflows stop being rebuilt per query: a pool plus a contextual bandit picks one, cutting MATH training tokens at least 40.55%
Workflow selection for LLM agents is recast as a contextual bandit over a reusable pool, reaching an 81.37% average across five benchmarks while cutting MATH training-phase token overhead by at least 40.55%.
Huawei Technologies (China)City University of Hong Kong
Jiahang Sun（First author）
CB-Orchestrator Contextual Bandit Agentic Workflow Workflow Optimization Evolutionary Search
arXiv Oct 12, 2026
A chest X-ray agent stops calling tools by default: it first asks whether evidence is needed, then blocks unusable measurements
The chest X-ray agent X-Ragent splits "should I seek evidence" from "is this measurement usable": the policy invokes region zoom, anatomical parsing and knowledge lookup only when needed, while quantitative measurements such as the cardiothoracic ratio enter reasoning only through an independent validity gate. Training runs in two stages, agentic supervised fine-tuning on 7,715 structured diagnostic trajectories followed by GRPO with an accuracy-gated reward that applies process terms only to correct trajectories. On the fully held-out ChestAgentBench, accuracy rises from 63.1% for the strongest baseline to 71.7%, with gains across all seven interpretation capabilities.
The chest X-ray agent X-Ragent splits "should I seek evidence" from "is this measurement usable": the policy invokes region zoom, anatomical parsing and knowledge lookup only when needed, while quantitative measurements such as the cardiothoracic ratio enter reasoning only through an independent validity gate. Training runs in two stages, agentic supervised fine-tuning on 7,715 structured diagnostic trajectories followed by GRPO with an accuracy-gated reward that applies process terms only to correct trajectories. On the fully held-out ChestAgentBench, accuracy rises from 63.1% for the strongest baseline to 71.7%, with gains across all seven interpretation capabilities.
Alibaba Group (China)National University of Singapore Beijing Jiaotong University City University of Hong Kong Santa Clara University Carnegie Mellon University
Yi Wang（First author）
X-Ragent Tool-Augmented Reinforcement Learning Rad-CoT GRPO Diagnostic Reasoning
arXiv Oct 8, 2026
Task-Progress Distillation Lets a 1.7B Agent Reach 72.4% Unseen Success on ALFWorld from 404 Demonstrations
The work introduces Task-Progress Distillation (TPD), an offline approach that pairs each demonstrated action with a short task-stage label so the student selects actions by jointly scoring admissible stage–action pairs; on ALFWorld, a 1.7B student trained with 404 demonstrations reaches 72.4% mean unseen task success with either TPD or action-only supervision, versus 48.3% for a reasoning-trained student using constrained action selection, and explicit stages raise success from 48.0% to 67.7% at 200 demonstrations, with action-only catching up to 76.9% at 808.
The work introduces Task-Progress Distillation (TPD), an offline approach that pairs each demonstrated action with a short task-stage label so the student selects actions by jointly scoring admissible stage–action pairs; on ALFWorld, a 1.7B student trained with 404 demonstrations reaches 72.4% mean unseen task success with either TPD or action-only supervision, versus 48.3% for a reasoning-trained student using constrained action selection, and explicit stages raise success from 48.0% to 67.7% at 200 demonstrations, with action-only catching up to 76.9% at 808.
City University of Hong Kong
Wenxi Gan（First author, Corresponding author）
Task-Progress Distillation ALFWorld Qwen3-1.7B LLM agent
arXiv Oct 6, 2026
ProsaBuddy uses LLM agents to prove real-time schedulability lemmas, significantly outperforming existing Rocq proving agents and the general coding agent OpenCode on a mini benchmark
The work presents ProsaBuddy, an LLM-based agent system that uses a ReAct loop with retrieval over the Prosa codebase, access to Rocq tools and optional human-written hints, and a subgoal-delegation architecture that decomposes a lemma into subgoals dispatched to subagents for proof, and on a mini benchmark drawn from real-time scheduling literature it significantly outperforms state-of-the-art LLM-based Rocq automated proving agent systems and the general coding agent OpenCode.
The work presents ProsaBuddy, an LLM-based agent system that uses a ReAct loop with retrieval over the Prosa codebase, access to Rocq tools and optional human-written hints, and a subgoal-delegation architecture that decomposes a lemma into subgoals dispatched to subagents for proof, and on a mini benchmark drawn from real-time scheduling literature it significantly outperforms state-of-the-art LLM-based Rocq automated proving agent systems and the general coding agent OpenCode.
Northeast Forestry University Northeastern University City University of Hong Kong Uppsala University Southeast University University of Electronic Science and Technology of China
Junyi Liu（First author）
ProsaBuddy ReAct loop Real-time schedulability analysis
arXiv Oct 5, 2026
CIM Decomposes MLLM Uncertainty via Causal-Invariant Masking, with EED Proxy Matching Performance at Nearly 50% Speedup
Addressing hallucinations in Multimodal Large Language Models (MLLMs), this work proposes Causal-Invariant Masking (CIM), which measures the semantic shift between original predictions and those conditioned on a causally-focused view to decompose uncertainty types, introduces Semantic Divergence as the core UQ metric with theoretical evidence that it converges to the variance of the model's sensitivity to non-causal correlations, and further proposes Expected Embedding Drift (EED), a fast geometric proxy metric; experiments report state-of-the-art performance on various benchmarks, with EED accelerating by nearly 50% at comparable performance.
Addressing hallucinations in Multimodal Large Language Models (MLLMs), this work proposes Causal-Invariant Masking (CIM), which measures the semantic shift between original predictions and those conditioned on a causally-focused view to decompose uncertainty types, introduces Semantic Divergence as the core UQ metric with theoretical evidence that it converges to the variance of the model's sensitivity to non-causal correlations, and further proposes Expected Embedding Drift (EED), a fast geometric proxy metric; experiments report state-of-the-art performance on various benchmarks, with EED accelerating by nearly 50% at comparable performance.
Southeast University The University of Sydney City University of Hong Kong Southeast University Apple (United States)
Haoyang Luo（First author）
Causal-Invariant Masking Semantic Divergence Expected Embedding Drift Uncertainty Quantification Multimodal Large Language Models