{"total":11,"items":[{"citing_arxiv_id":"2606.21022","ref_index":45,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Structure-Aware Graph Multi-Task Learning for Dynamic Sparse OD Demand Prediction","primary_cat":"cs.LG","submitted_at":"2026-06-19T01:17:48+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"SAGMTL decomposes dynamic sparse OD demand prediction into joint structural state modeling and flow intensity estimation via node-edge collaborative graph representations.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.22611","ref_index":73,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Benchmarking Machine Learning Architectures for Antimicrobial Stewardship in Pediatric ICUs","primary_cat":"cs.LG","submitted_at":"2026-05-21T15:26:58+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Benchmarking in pediatric ICU antimicrobial stewardship shows performance depends mainly on target prevalence and dataset traits rather than model complexity, with sequence models improving precision-recall at 24-hour resolution but showing poorer calibration than tabular models.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.17624","ref_index":3,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Multi-task learning on partially labeled datasets via invariant/equivariant semi-supervised learning","primary_cat":"cs.CV","submitted_at":"2026-05-17T19:43:05+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"Invariant and equivariant semi-supervised learning improves multi-task detection and segmentation performance on partially labeled vision datasets compared to supervised baselines.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.16991","ref_index":6,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Response-free item difficulty modelling for multiple-choice items with fine-tuned transformers: Component-wise representation and multi-task learning","primary_cat":"cs.CL","submitted_at":"2026-05-16T13:22:57+00:00","verdict":"CONDITIONAL","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Fine-tuned transformers with multi-task learning recover substantial wording-derived signal for item difficulty at small sample sizes typical in applied testing.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"[CLS] is the [CLS] hidden state of the encoder's output for that input. Each pooled representation is passed through a classification head fξ that produces a scalar \"logit\". The function fξ has the same form as gψ of Equation (2), with its own parameters collected under ξ={u∈R d, e∈R} , shared across all M options (input sequences): zi,m =f ξ r(optm) i \u0001 =u ⊤r(optm) i +eform= 1, . . . , M .(6) What distinguishes the two heads is the task they serve and the loss they are paired with: gψ outputs a scalar prediction trained under the MSE loss (Equation 3); fξ outputs per-option \"logits\" zi,m and its parameters ξ are trained using the cross-entropy loss defined below (Equation 7). Keeping fξ as a single linear layer devotes the discriminative capacity"},{"citing_arxiv_id":"2605.06231","ref_index":32,"ref_count":2,"confidence":0.88,"is_internal_anchor":false,"paper_title":"YEZE at SemEval-2026 Task 9: Detecting Multilingual, Multicultural and Multievent Online Polarization via Heterogeneous Ensembling","primary_cat":"cs.CL","submitted_at":"2026-05-07T13:21:40+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":2.0,"formal_verification":"none","one_line_summary":"A heterogeneous ensemble of XLM-RoBERTa-large and mDeBERTa-v3-base with independent task modeling and class weighting is reported as effective for multilingual, multicultural, and multievent online polarization detection.","context_count":1,"top_context_role":"method","top_context_polarity":"use_method","context_text":"imbalanced settings. Such models are typically based on Transformer architectures (Vaswani et al., 2017), with pre- trained variants such as BERT (Devlin et al., 2019) offering strong contextual representations for down- stream classification tasks. Within this paradigm, a common strategy is to jointly model related objectives using multi-task learning (MTL) (Caruana, 1997). However, prior work has shown that MTL can suffer from negative transfer and task interference, particularly when label distributions are highly imbalanced or hetero- geneous (Yu et al., 2020). In contrast, we adopt independent task modeling combined with ensem- ble learning, which proves to be more robust under severe label sparsity. 3 System Overview"},{"citing_arxiv_id":"2605.05851","ref_index":87,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Hypothesis generation and updating in large language models","primary_cat":"cs.LG","submitted_at":"2026-05-07T08:24:54+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"LLMs exhibit Bayesian-like hypothesis updating with strong-sampling bias and an evaluation-generation gap but generalize poorly outside observed data.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.26375","ref_index":6,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"SG-UniBuc-NLP at SemEval-2026 Task 6: Multi-Head RoBERTa with Chunking for Long-Context Evasion Detection","primary_cat":"cs.CL","submitted_at":"2026-04-29T07:37:03+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":3.0,"formal_verification":"none","one_line_summary":"A multi-head RoBERTa model with overlapping chunking and max-pooling achieves Macro-F1 of 0.80 on 3-way clarity classification and 0.51 on 9-way evasion strategy detection, ranking 11th in both subtasks of SemEval-2026 Task 6.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.20268","ref_index":2,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Opportunistic Bone-Loss Screening from Routine Knee Radiographs Using a Multi-Task Deep Learning Framework with Sensitivity-Constrained Threshold Optimization","primary_cat":"cs.CV","submitted_at":"2026-04-22T07:12:04+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":4.0,"formal_verification":"none","one_line_summary":"STR-Net achieves AUROC of 0.933 for binary bone-loss screening and 0.801 correlation for T-score estimation from knee X-rays on a held-out test set.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.17698","ref_index":70,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"The Geometric Canary: Predicting Steerability and Detecting Drift via Representational Stability","primary_cat":"cs.LG","submitted_at":"2026-04-20T01:24:45+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Task-aligned supervised geometric stability predicts linear steerability with high accuracy while unsupervised stability detects representational drift earlier and with lower false alarms than CKA or Procrustes.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.05271","ref_index":6,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Toward Unified Fine-Grained Vehicle Classification and Automatic License Plate Recognition","primary_cat":"cs.CV","submitted_at":"2026-04-07T00:10:56+00:00","verdict":"ACCEPT","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"UFPR-VeSV is a new real-world dataset for fine-grained vehicle classification and automatic license plate recognition collected from Brazilian police cameras, with benchmarks demonstrating its difficulty and the value of joint task use.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2110.02879","ref_index":5,"ref_count":1,"confidence":0.88,"is_internal_anchor":false,"paper_title":"Compositional Q-learning for electrolyte repletion with imbalanced patient sub-populations","primary_cat":"cs.LG","submitted_at":"2021-10-06T16:08:05+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"CFQI extends fitted Q-iteration by using separate modules for compositional task variants to learn policies robust to imbalanced patient sub-populations in medical RL.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}