Skip to main content
Search

Southern University of Science and Technology

16 related articles

Subscribe to updates
  1. arXiv

    ForceWAM splits contact learning in two: demos set the goal, play data teach force response, reaching 95% on one real-robot task

    In contact-rich manipulation, expert demonstrations cover successful paths but say little about how force responds to different actions, while task-agnostic play records broader contact dynamics without a task goal. ForceWAM gives the task-compatible motion and six-axis force/torque future to a force world model trained on demonstrations, and gives the question of which actions realize that future to an inverse dynamics model trained on both demonstrations and play; at deployment, fresh wrench feedback corrects actions between world-model updates. Across force-regulated sliding, box flipping with and without disturbances, and ring placement, the full system reaches 85%, 95%, 85%, and 90% success, above the listed baselines.

  2. arXiv

    Retinal cross-modal registration skips segmentation: MICRO benchmark and one-step OFS2Reg reach 0.784 success rate

    Cross-modal retinal image rigid registration has largely relied on a two-step pipeline that segments vessels first and detects keypoints second, so registration quality is capped by segmentation accuracy. The MICRO benchmark gathers 1,365 cross-modal retinal image pairs spanning preoperative, intraoperative and postoperative surgery phases, each with expert-annotated control points. Built on that benchmark, the one-step OFS2Reg constructs keypoint topology constraints directly through a frequency-guided structure preservation descriptor and cross-modal vessel structure similarity learning, reaching a 0.784 success rate and 5.08-pixel RMSE on the MICRO-pre subset.

  3. bioRxiv

    Folding models retooled as antibody screeners: frozen Protenix lifts Origin-1 macro AUROC from 0.669 to 0.757

    Antibody screening usually splits specificity, affinity and developability across separate models, and structural information is often compressed into a single confidence score such as ipTM. FoldAbS freezes the open-source folding model Protenix as a shared encoder, trains only lightweight task heads, and supplies specificity supervision with pseudo-negatives built from antigen mismatch, CDRH3 mutation and CDRH3 replacement. With Protenix as backbone, FoldAbS-Px raises macro AUROC on Origin-1 from 0.6690 to 0.7570 and reports the best results on affinity and four developability properties.

  4. arXiv

    Preference optimization gains a linguistic constraint: folding information uniformity into the reward brings model wording closer to human habits

    A preference optimization reward built on the psycholinguistic uniform information density hypothesis matches SimPO on AlpacaEval 2, Arena-Hard and MT-Bench, and moves English complement-clause that usage closer to natural language.

  5. arXiv

    Information events in language have predictable arrival times: new-word intensity decays, and LLMs introduce new words faster than humans

    Treating natural language as a temporal point process of information-bearing events lets a neural model predict when the next new word or high-surprisal token will arrive, and shows LLMs introducing new words at consistently higher rates than humans.

  6. bioRxiv

    Phage matching stops relying on one embedding: routing by query novelty lifts both-unseen AUPRC from 0.036 to 0.671

    Phage therapy needs a matching phage for each clinical bacterial isolate, and conventional experimental screening is slow and labor-intensive. PhageHost-GNN encodes phage receptor-binding proteins and bacterial K-locus proteins as sequence, structure and genomic-context embeddings, feeds them to a graph neural network, and automatically selects the embedding model best suited to how novel a query strain is relative to the training set. On the Spain-derived Beamud2023 data, with both hosts and phages unseen, the routed model reached an AUPRC of 0.671, while fixed-embedding models stayed at or below 0.036.

  7. arXiv

    SteerScope benchmark: none of 23 steering methods beats the Prompt Steering baseline on the efficacy–side-effect trade-off

    The authors introduce SteerScope, a two-axis, 15-metric evaluation suite for LLM activation steering, and benchmark 23 methods spanning 4 families (including prompting, LoRA, and SFT baselines) on Gemma-2-2B-it and Gemma-2-9B-it under matched models, tasks, and protocols; they find that no activation steering method achieves higher target efficacy without greater composite side effects, so none surpasses the Prompt Steering baseline on the efficacy–side-effect trade-off, and that efficacy and side effects rise together with intervention strength.

  8. arXiv

    Representation geometry explains why forgotten knowledge is easily relearned, motivating an MCU unlearning method that targets minor components

    This work studies, from a representation geometry perspective, why unlearned models rapidly recover forgotten knowledge under relearning attacks, finding that existing unlearning methods mainly optimize along dominant components while leaving minor components largely unchanged, that modifications in dominant components are easily reversed during relearning whereas minor components resist such reversal, and proposes Minor Component Unlearning (MCU), which explicitly targets minor components and is validated on three datasets to substantially improve resistance to relearning attacks.

  9. arXiv

    CATCH testbed makes reward hacking in coding RL reproducible and shows chain-of-thought monitoring erodes as policies mislead it with code comments

    The authors introduce CATCH, a controllable testbed for reward hacking in coding RL that deliberately exposes environmental loopholes and provides execution-based gold labels by comparing success under a vulnerable evaluator with task correctness under an independent audit, while controlling the model's initial hacking tendency through supervised fine-tuning data mixtures and the difficulty of earning rewards through reward designing; experiments show CATCH produces diverse RL training trajectories with clear reward hacking, that initial models and reward difficulties jointly shape its emergence, and that a chain-of-thought monitor initially suppresses hacking but this protection erodes as the policy model learns to mislead the monitor with code comments.

  10. Two-center study of 127 nuclear cataract cases: AS-OCT image features correlate with LOCS III grading, with automated grading accuracy of 0.81 and 0.87

    This two-center clinical validation study recruited 127 individuals with different severities of nuclear cataract from Thailand (n = 81) and Shenzhen, China (n = 46), imaged them with AS-OCT and graded images under the LOCS III standard, developed automated machine learning models to extract nuclear region annotation and analyzed feature-based quantifiers, finding that pixel-based features such as mean, variance, root mean square, interquartile range, and percentiles significantly correlate with NC grading (P < .01), that variance, standard deviation, and median showed high consistency while kurtosis and skewness were negatively correlated, and that the prediction model achieved 0.81 accuracy at the SZRM center (F1 0.82) and 0.87 at the Thai center (F1 0.